Oral structure information generation method, system, storage medium and oral device

Abstract

The present invention discloses a method, system, storage medium and oral device for generating oral structure information. The method comprises: obtaining oral three-dimensional model data, judging whether the intraoral tissue area specified in the oral three-dimensional model satisfies a preset integrity condition; if not, obtaining corresponding intraoral photo image data, extracting oral feature data in the intraoral photo image data, establishing a pixel size relationship between the intraoral photo image data and the oral three-dimensional model data, and reconstructing a part representing the oral feature data on the oral three-dimensional model data to obtain oral structure information. The method for generating oral structure information provided by the present invention can simplify logic, improve data integrity and accuracy, and obtain more intuitive and accurate structure information.

Description

Technical Field

[0001] The present invention relates to the technical field of oral digital models, and in particular to an oral structure information generation method, system, storage medium and oral appliance. Background Art

[0002] In today's society, oral health has gradually become a focus of people's attention. Oral health not only includes oral hygiene achieved by controlling plaque, removing dirt and food residues, but also includes oral health maintenance represented by improving occlusal relationship and orofacial muscle function. For the latter, on the one hand, it will affect dental hygiene, chewing function, the development and function of the temporomandibular joint, and easily cause malocclusion and snoring, which is a double hazard to human health and aesthetics; on the other hand, the adjustment of occlusal relationship and orofacial muscle function depends on high-precision analysis of oral structure information, especially the characteristics of intraoral tissue structure. When there are large errors, it will cause problems such as poor treatment effect and user experience. It can be seen that in order to effectively maintain oral health, the analysis and extraction of oral structure information is crucial.

[0003] A common technical solution in the prior art is to make a silicone model of the patient's oral cavity and use it to further make an oral plaster model or an oral three-dimensional model, or to use a scanner to scan the oral cavity to obtain a patient's oral scan model, so as to analyze the oral structure information. However, during the extraction process of the above model, on the one hand, too much attention will be paid to the shape and position of the crown, resulting in the loss of other features in the oral cavity, and it is difficult to provide comprehensive data reference when conducting a comprehensive diagnosis; on the other hand, considering operational errors and patient comfort, it is very likely that regional features will be incompletely collected, and the model needs to be re-made or collected for local features. In addition, because the steps of making an oral silicone model or an oral scan model are cumbersome, simply using intraoral images cannot meet the accuracy requirements. Based on this, it is necessary to provide a method for generating oral structure information to solve the above problems. Summary of the invention

[0004] One of the purposes of the present invention is to provide a method for generating oral structure information to solve the problem that the oral feature analysis step in the prior art cannot take into account the advantages of comprehensiveness of feature data, high precision requirements and patient comfort, and the degree of step simplification, and cannot realize the automation of judgment logic and information generation.

[0005] One of the objectives of the present invention is to provide an oral structure information generation system.

[0006] One of the objectives of the present invention is to provide a storage medium.

[0007] One of the objects of the present invention is to provide an oral appliance.

[0008] To achieve one of the above-mentioned purposes of the invention, an embodiment of the present invention provides a method for generating oral structure information, comprising: obtaining oral three-dimensional model data, determining whether an intraoral tissue area specified in the oral three-dimensional model satisfies a preset integrity condition; if not, obtaining corresponding intraoral photo image data, extracting oral feature data corresponding to the intraoral tissue area in the intraoral photo image data, establishing a pixel size relationship between the intraoral photo image data and the oral three-dimensional model data, and reconstructing a portion representing the oral feature data on the oral three-dimensional model data to obtain the oral structure information.

[0009] As a further improvement of an embodiment of the present invention, the "obtaining oral three-dimensional model data, and determining whether the intraoral tissue area specified in the oral three-dimensional model satisfies a preset integrity condition" specifically includes: obtaining oral three-dimensional model data, and calculating the positional relationship between the target tooth position and the model boundary and / or the positional relationship between the maxillary model boundary and the mandibular model boundary according to the target tooth position features in the oral three-dimensional model data and the model boundary features corresponding to the target tooth position features; comparing the positional relationship with the preset position condition corresponding to the positional relationship, and determining whether the intraoral tissue area corresponding to the target tooth position in the oral three-dimensional model satisfies the preset integrity condition.

[0010] As a further improvement of an embodiment of the present invention, the "calculating the positional relationship between the target tooth position and the model boundary according to the target tooth position feature in the oral three-dimensional model data and the model boundary feature corresponding to the target tooth position feature" specifically includes: determining at least one tooth position on the oral three-dimensional model as the target tooth position, and calculating the reference feature coordinates of the target tooth position based on a preset feature recognition rule to characterize the target tooth position feature; determining the boundary feature coordinates corresponding to the reference feature coordinates on the oral three-dimensional model along a first direction to characterize the model boundary feature; calculating the feature distance value between the reference feature coordinates and the boundary feature coordinates to characterize the positional relationship between the target tooth position and the model boundary; the "comparing the positional relationship with the preset position condition corresponding to the positional relationship to determine whether the intraoral tissue area corresponding to the target tooth position in the oral three-dimensional model meets the preset integrity condition" specifically includes: comparing the numerical value of the feature distance value with the distance integrity criterion value characterizing the preset position condition, and judging whether the intraoral tissue area corresponding to the target tooth position in the oral three-dimensional model meets the preset integrity condition according to the comparison result.

[0011] As a further improvement of an embodiment of the present invention, the reference feature coordinates are located in a preset model coordinate system, and the origin of the model coordinate system is located in the occlusal plane; the "calculating the reference feature coordinates of the target tooth position based on a preset feature recognition rule" specifically includes: determining all spatial points on the crown of the target tooth position located on the labial side to form a set of spatial points; selecting the spatial point with the smallest coordinate value in the set of spatial points as the pole to establish a polar coordinate system, and arranging the other spatial points in ascending order according to the polar angle and polar diameter to form a feature traversal sequence; extracting the coordinates of the first two spatial points in the feature traversal sequence, taking the pole as the starting point, successively calculating the first polar coordinate vector and the second polar coordinate vector, and calculating the first polar coordinate vector. The cross product of the scalar vector and the second polar coordinate vector is determined to determine whether the cross product is less than 0; if so, the starting point is updated to be the first spatial point corresponding to the first polar coordinate vector; if not, the starting point is updated to be the second spatial point corresponding to the second polar coordinate vector; all other spatial points in the feature traversal sequence that are located after the second spatial point are traversed, and according to the polar coordinate vector formed by the spatial point and the starting point, and the cross product between the polar coordinate vectors, the starting point is selectively updated and at least two spatial points that meet the judgment conditions are determined to be convex hull reference points; according to the convex hull reference point and the pole, the convex hull contour data of the target tooth position is fitted, and according to the convex hull contour data, the reference feature coordinates of the target tooth position are calculated.

[0012] As a further improvement of an embodiment of the present invention, the "traversing all other spatial points in the feature traversal sequence that are located after the second spatial point, updating the starting point and determining at least two spatial points that meet the judgment conditions as convex hull reference points according to the polar coordinate vector formed by the spatial point and the starting point, and the cross product between the polar coordinate vectors" specifically includes: extracting a third spatial point that is located after the second spatial point in the feature traversal sequence, calculating in turn the reference polar coordinate vector between the reference spatial point that is not the starting point in the first spatial point and the second spatial point and the starting point, and the third polar coordinate vector between the third spatial point and the starting point vector, and calculate the cross product of the reference polar coordinate vector and the third polar coordinate vector, and determine whether the cross product is less than 0; if so, update the starting point to the reference space point, and determine the reference space point to be the convex hull reference point; if not, do not update the starting point, delete the reference space point, use the third space point as the new reference space point, and based on the new reference polar coordinate vector and the cross product of the polar coordinate vector formed by the starting point and the next space point of the new reference space point, selectively update the starting point and determine the convex hull reference point; repeat the iteration until the judgment of all space points in the feature traversal sequence is completed, and at least two convex hull reference points are obtained.

[0013] As a further improvement of an embodiment of the present invention, the "calculating the reference feature coordinates of the target tooth position based on preset feature recognition rules" specifically includes: determining and using the coordinates of the midpoint of the gingival margin of the target tooth position as the reference feature coordinates of the target tooth position.

[0014] As a further improvement of an embodiment of the present invention, the "determining, according to the reference feature coordinates, boundary feature coordinates on the oral three-dimensional model corresponding to the reference feature coordinates along the first direction to characterize the boundary features of the model" specifically includes: taking the reference feature coordinates as the starting point, making a reference extension line along the first direction away from the target tooth position, and continuously analyzing the formation of the projection endpoints of the reference extension ends on the reference extension line except the reference feature coordinates on the maxillary surface of the oral three-dimensional model; when the reference extension end no longer forms a projection endpoint with the maxillary surface of the oral three-dimensional model, determining the last formed projection endpoint coordinates as the boundary feature coordinates.

[0015] As a further improvement of one embodiment of the present invention, the intraoral tissue area includes the vestibule groove, the oral feature data includes the vestibule groove height data, the first direction is the projection of the extension direction of the long axis of the tooth on the maxillary surface of the oral three-dimensional model, and the target tooth position includes the maxillary incisor position and the mandibular incisor position.

[0016] As a further improvement of an embodiment of the present invention, the intraoral tissue area includes the vestibule groove and / or the root ridge, the oral feature data includes dental arch width data, the first direction is the projection of the extension direction of the long axis of the tooth on the maxillary surface of the oral three-dimensional model, and the target tooth position includes the left molar tooth position of the first maxillary surface and the right molar tooth position of the first maxillary surface corresponding to each other; wherein, the first maxillary surface is the maxilla and / or mandible.

[0017] As a further improvement of an embodiment of the present invention, the intraoral tissue area includes a lip frenulum, the oral feature data includes lip frenulum width data, and the "determining at least one tooth position on the oral three-dimensional model as the target tooth position, and calculating the reference feature coordinates of the target tooth position based on a preset feature recognition rule to characterize the target tooth position characteristics" specifically includes: determining the left central incisor tooth position of the first maxillary face and the right central incisor tooth position of the first maxillary face on the oral three-dimensional model as the target tooth position, and calculating the reference feature coordinates of the target tooth position based on a preset feature recognition rule, as well as a number of relative feature coordinates on the target tooth position located between the two reference feature coordinates to characterize the target tooth position characteristics; wherein the first maxillary face is the maxilla and / or mandible; "According to the reference feature coordinates, determine the boundary feature coordinates on the oral three-dimensional model corresponding to the reference feature coordinates along the first direction to characterize the boundary features of the model", specifically including: taking the reference feature coordinates and the relative feature coordinates as starting points, respectively, making reference extension lines along the first direction away from the target tooth position, and continuously analyzing the formation of projection endpoints of the reference extension ends on the reference extension lines other than the reference feature coordinates or the relative feature coordinates on the oral three-dimensional model; wherein, the first direction is the projection of the extension direction of the long axis of the tooth on the maxillary surface of the oral three-dimensional model; when the reference extension end no longer forms the projection endpoint with the oral three-dimensional model, determining the last formed projection endpoint coordinates as the boundary feature coordinates.

[0018] As a further improvement of an embodiment of the present invention, before the step of "comparing the characteristic distance value with the distance integrity criterion value characterizing the preset position condition, and judging whether the intraoral tissue area corresponding to the target tooth position in the oral three-dimensional model meets the preset integrity condition based on the comparison result", it includes: acquiring at least two sets of three-dimensional training model data, and at least two sets of distance training data corresponding to the three-dimensional training model data; wherein the distance training data includes the conditional training coordinates corresponding to the reference characteristic coordinates on the three-dimensional training model, and the distance between the target tissue coordinates corresponding to the intraoral tissue area on the three-dimensional training model; the line connecting the conditional training coordinates and the target tissue coordinates extends along the first direction; calculating the average distance data and the training distance standard deviation of the distance training data, and calculating the distance integrity criterion value based on the difference between the product of the average distance data and the training distance standard deviation and the preset distribution probability coefficient.

[0019] As a further improvement of an embodiment of the present invention, the intraoral tissue area includes a vestibule groove, the oral feature data includes vestibule groove height data, the target tooth position includes a maxillary first incisor tooth position and a mandibular second incisor tooth position, and the distance training data includes a first groove bottom distance parameter corresponding to the maxillary first incisor tooth position, and a second groove bottom distance parameter corresponding to the mandibular second incisor tooth position; the "calculating the average distance data and training distance standard deviation of the distance training data" specifically includes: calculating the first groove bottom average distance value of all first groove bottom distance parameters in all distance training data, and the second groove bottom average distance value of all second groove bottom distance parameters in all distance training data, to obtain the average distance data; calculating the first groove bottom distance standard deviation of all first groove bottom distance parameters in all distance training data, and the second groove bottom distance standard deviation of all second groove bottom distance parameters in all distance training data, to obtain the training distance standard deviation.

[0020] As a further improvement of an embodiment of the present invention, the distance training data includes groove bottom distance parameters corresponding to the maxillary left lateral incisor tooth position, the maxillary left central incisor tooth position, the maxillary right central incisor tooth position, the maxillary right lateral incisor tooth position, the mandibular left lateral incisor tooth position, the mandibular left central incisor tooth position, the mandibular right central incisor tooth position and the mandibular right lateral incisor tooth position; the groove bottom distance parameters represent the distance between the conditional training coordinates of the midpoint of the gingival margin of the corresponding incisor tooth position and the target tissue coordinates of the groove bottom feature point corresponding to the midpoint of the gingival margin.

[0021] As a further improvement of an embodiment of the present invention, the intraoral tissue area includes a vestibule groove and / or a root ridge, the oral feature data includes dental arch width data, the target tooth position includes distal molar positions on both sides of the maxilla and distal molar positions on both sides of the mandible, and the distance training data includes upper dental arch width parameters corresponding to the distal molar positions on both sides of the maxilla, and lower dental arch width parameters corresponding to the distal molar positions on both sides of the mandible; the "calculating the average distance data and training distance standard deviation of the distance training data" specifically includes: calculating the upper average width value of all upper dental arch width parameters in all distance training data, and the lower average width value of all lower dental arch width parameters in all distance training data, to obtain the average distance data; calculating the upper width standard deviation of all upper dental arch width parameters in all distance training data, and the lower width standard deviation of all lower dental arch width parameters in all distance training data, to obtain the training distance standard deviation.

[0022] As a further improvement of an implementation manner of the present invention, the distribution probability coefficient is 2.

[0023] As a further improvement of an embodiment of the present invention, the "calculating the characteristic distance value between the reference feature coordinates and the boundary feature coordinates" specifically includes: calculating the distance value between the reference feature coordinates and the boundary feature coordinates in the extension direction of the tooth centerline to obtain the characteristic distance value.

[0024] As a further improvement of an embodiment of the present invention, the first direction is the projection of the extension direction of the long axis of the tooth on the three-dimensional model of the oral cavity; the boundary feature coordinates and the reference feature coordinates are located in a preset model coordinate system, and the model coordinate system at least includes a first coordinate axis and a third coordinate axis, the third coordinate axis extends along the extension direction of the tooth midline, and the first coordinate axis extends along the extension direction of the width of the central incisor; the "calculating the distance value between the reference feature coordinates and the boundary feature coordinates in the extension direction of the tooth midline" specifically includes: calculating the coordinate difference between the reference feature coordinates and the boundary feature coordinates on the third coordinate axis.

[0025] As a further improvement of an embodiment of the present invention, the "comparing the numerical values ​​of the characteristic distance value with the distance integrity criterion value characterizing the preset position condition, and judging whether the intraoral tissue area corresponding to the target tooth position in the oral three-dimensional model satisfies the preset integrity condition based on the comparison result" specifically includes: judging whether the characteristic distance value is less than the distance integrity criterion value; if so, judging that the position in the oral three-dimensional model pointed to by the characteristic distance value is a characteristic missing position; if not, continuing to compare the numerical values ​​of the next characteristic distance value with the distance integrity criterion value; judging whether the intraoral tissue area corresponding to the target tooth position in the oral three-dimensional model satisfies the integrity condition based on the number of the characteristic missing positions.

[0026] As a further improvement of an embodiment of the present invention, the "determining whether the intraoral tissue area corresponding to the target tooth position in the oral three-dimensional model satisfies the integrity condition based on the number of the feature missing positions" specifically includes: determining the numerical relationship between the number of the feature missing positions and the allowable error quantity value; if the number of the feature missing positions is greater than or equal to the allowable error quantity value, determining that the intraoral tissue area corresponding to the target tooth position in the oral three-dimensional model does not satisfy the integrity condition; if the number of the feature missing positions is less than the allowable error quantity value, determining that the intraoral tissue area corresponding to the target tooth position in the oral three-dimensional model satisfies the integrity condition; wherein the allowable error quantity value is an integer greater than or equal to one half of the number of target tooth positions.

[0027] As a further improvement of an embodiment of the present invention, the "calculating the positional relationship between the maxillary model boundary and the mandibular model boundary according to the target tooth position features in the oral three-dimensional model data and the model boundary features corresponding to the target tooth position features" specifically includes: determining at least the incisor tooth position in the oral three-dimensional model as the target tooth position, and calculating the reference feature coordinates of the target tooth position based on a preset feature recognition rule to characterize the target tooth position features; determining the upper boundary extreme value coordinates and the lower boundary extreme value coordinates located in the oral three-dimensional model along a first direction according to the reference feature coordinates; wherein the upper boundary extreme value coordinates are located in the maxilla of the oral three-dimensional model, and the distance between the upper boundary extreme value coordinates and the occlusal plane has a maximum value, and the lower boundary extreme value coordinates The lower jaw is located in the oral three-dimensional model, and the distance between it and the occlusal plane has a maximum value; the characteristic height value between the upper boundary extreme value coordinates and the lower boundary extreme value coordinates is calculated to characterize the positional relationship between the maxillary model boundary and the mandibular model boundary; the "comparing the positional relationship with the preset position condition corresponding to the positional relationship, and judging whether the intraoral tissue area corresponding to the target tooth position in the oral three-dimensional model meets the preset integrity condition" specifically includes: comparing the numerical values ​​of the characteristic height value with the height integrity criterion value characterizing the preset position condition, and judging whether the intraoral tissue area corresponding to the target tooth position in the oral three-dimensional model meets the preset integrity condition based on the comparison result.

[0028] As a further improvement of an embodiment of the present invention, the first direction is the projection of the extension direction of the long axis of the tooth on the maxillary surface of the oral three-dimensional model, the reference feature coordinates, the upper boundary extreme value coordinates and the lower boundary extreme value coordinates are located in a preset model coordinate system, the origin of the model coordinate system is located on the straight line where the incisor ridge is located, and at least includes a first coordinate axis and a third coordinate axis, the third coordinate axis extends along the extension direction of the tooth midline, and the first coordinate axis extends along the extension direction of the central incisor width; the "determining the upper boundary extreme value coordinates and the lower boundary extreme value coordinates located in the oral three-dimensional model along the first direction according to the reference feature coordinates" specifically includes: calculating the coordinates of the upper boundary extreme value and the lower boundary extreme value on the oral three-dimensional model. A set of upper boundary coordinates corresponding to the maxillary target tooth position, and a set of lower boundary coordinates corresponding to the mandibular target tooth position on the oral three-dimensional model, to characterize the boundary features of the model; traverse and determine the upper boundary coordinate with the maximum coordinate value corresponding to the third coordinate axis in the upper boundary coordinate set as the upper boundary extreme value coordinate, traverse and determine the lower boundary coordinate with the minimum coordinate value corresponding to the third coordinate axis in the lower boundary coordinate set as the lower boundary extreme value coordinate; the "calculating the characteristic height value between the upper boundary extreme value coordinate and the lower boundary extreme value coordinate" specifically includes: calculating the coordinate difference between the upper boundary extreme value coordinate and the lower boundary extreme value coordinate on the third coordinate axis as the characteristic height value.

[0029] As a further improvement of an embodiment of the present invention, before the "comparing the numerical values ​​of the characteristic height value with the height integrity criterion value characterizing the preset position condition, and judging whether the intraoral tissue area corresponding to the target tooth position in the oral three-dimensional model meets the preset integrity condition based on the comparison result", it includes: obtaining at least two sets of three-dimensional training model data, and at least two sets of height training data corresponding to the three-dimensional training model data; wherein, the height training data includes the distance between the upper tissue extreme value coordinates on the three-dimensional training model corresponding to the intraoral tissue area on one side of the first direction and the lower tissue extreme value coordinates on the three-dimensional training model corresponding to the intraoral tissue area on the other side of the first direction; calculating the average height data and the training height standard deviation of the height training data, and calculating the height integrity criterion value according to the difference between the product of the average height data and the training height standard deviation and the preset distribution probability coefficient; wherein, the distribution probability coefficient is 3.

[0030] As a further improvement of one embodiment of the present invention, the "extracting oral feature data corresponding to the intraoral tissue area in the intraoral image data" specifically includes: calling the corresponding first neural network model and second neural network model according to the intraoral tissue area; inputting the intraoral image data into the first neural network model to extract the region of interest to obtain feature area image data; inputting the feature area image data into the second neural network model to perform feature recognition to obtain the oral feature data corresponding to the intraoral tissue area.

[0031] As a further improvement of an embodiment of the present invention, the first neural network model and the second neural network model are convolutional neural network models; the intraoral tissue area includes the vestibule sulcus; before the "calling the corresponding first neural network model and the second neural network model according to the intraoral tissue area", it includes: obtaining at least two sets of training image data, and region of interest marks corresponding to the training image data; wherein the extension range of the region of interest marks in the corresponding training image at least covers the upper lip, lower lip and dentition in the training image; calling a preset convolutional neural network model, and performing iterative training with the training image data and the region of interest marks as model inputs to obtain first training parameters and a corresponding first neural network model; obtaining at least two sets of training image data, and tooth position marks and vestibule sulcus bottom marks corresponding to the training image data; calling a preset convolutional neural network model, and performing iterative training with the training image data, the tooth position marks and the vestibule sulcus bottom marks as model inputs to obtain second training parameters and a corresponding second neural network model.

[0032] As a further improvement of an embodiment of the present invention, the "establishing a pixel size relationship between the intraoral image data and the oral three-dimensional model data" specifically includes: determining at least one target tooth position corresponding to each other in the intraoral image and the oral three-dimensional model as a relative reference tooth position, and calculating the size data of the crown of the relative reference tooth position in at least one same direction in the intraoral image and the oral three-dimensional model, and respectively obtaining reference pixel size data and reference physical size data; according to the reference pixel size data and the reference physical size data, fitting a size mapping factor to characterize the pixel size relationship.

[0033] As a further improvement of an embodiment of the present invention, the at least one same direction includes a crown width direction; and the size mapping factor is a quotient of the reference physical size data and the reference pixel size data.

[0034] As a further improvement of an embodiment of the present invention, the "calculating the dimension data of the crown of the relative reference tooth position in at least one same direction in the intraoral image and the oral three-dimensional model, and obtaining the reference pixel dimension data and the reference physical dimension data respectively" specifically includes: determining the first reference feature point and the second reference feature point of the relative reference tooth position on the intraoral image, calculating and obtaining the reference pixel dimension data based on the number of pixels between the first reference feature point and the second reference feature point; wherein the first reference feature point and the second reference feature point are located on both sides of the long axis of the tooth of the relative reference tooth position, and the first reference feature point The distance from the corresponding relative reference tooth position crown incisal end is equal to the distance between the second reference feature point and the corresponding relative reference tooth position crown incisal end; the third reference feature point and the fourth reference feature point of the relative reference tooth position are determined on the oral three-dimensional model, and the Euclidean distance between the third reference feature point and the fourth reference feature point is calculated and obtained to obtain the reference physical dimension data; wherein, the third reference feature point and the fourth reference feature point are located on both sides of the long axis of the tooth body of the relative reference tooth position, and the distance between the third reference feature point and the corresponding relative reference tooth position crown incisal end is equal to the distance between the fourth reference feature point and the corresponding relative reference tooth position crown incisal end.

[0035] As a further improvement of an embodiment of the present invention, the "determining at least one target tooth position corresponding to each other in the intraoral image and the oral three-dimensional model as a relative reference tooth position" specifically includes: judging whether the maxillary left central incisor tooth position and the maxillary right central incisor tooth position in the intraoral image and the oral three-dimensional model both contain complete tooth crowns; if so, determining the maxillary left central incisor tooth position and the maxillary right central incisor tooth position corresponding to each other in the intraoral image and the oral three-dimensional model as relative reference tooth positions; the "determining the first reference feature point and the second reference feature point of the relative reference tooth position on the intraoral image, calculating and obtaining the reference pixel size data based on the number of pixels between the first reference feature point and the second reference feature point" specifically includes: determining the first reference feature point at the distal boundary of the crown of the maxillary left central incisor tooth position in the intraoral image, and determining the second reference feature point at the distal boundary of the crown of the maxillary right central incisor tooth position in the intraoral image; calculating the number of pixels between the first reference feature point and the second reference feature point, and taking half of the number of pixels as the reference pixel size data.

[0036] As a further improvement of an embodiment of the present invention, the distance from the first reference feature point to the gingival end of the corresponding incisor tooth position is three times the distance from the first reference feature point to the incisal end of the crown of the corresponding incisor tooth position.

[0037] As a further improvement of an embodiment of the present invention, the "determining at least one target tooth position corresponding to each other in the intraoral image and the oral three-dimensional model as a relative reference tooth position" specifically includes: judging whether the maxillary left central incisor tooth position and the maxillary right central incisor tooth position in the intraoral image and the oral three-dimensional model both contain complete tooth crowns; if only the maxillary left central incisor tooth position in the intraoral image and the oral three-dimensional model contains a complete tooth crown, then determining the maxillary left central incisor tooth position corresponding to each other in the intraoral image and the oral three-dimensional model as the relative reference tooth position; ... The method comprises: determining a first reference feature point and a second reference feature point of the relative reference tooth position, and calculating and obtaining the reference pixel size data according to the number of pixels between the first reference feature point and the second reference feature point, specifically comprising: determining the first reference feature point at the distal boundary of the crown of the maxillary left central incisor tooth position in the intraoral image, and determining the second reference feature point at the mesial boundary of the crown of the maxillary left central incisor tooth position in the intraoral image; calculating the number of pixels between the first reference feature point and the second reference feature point, and using the number of pixels as the reference pixel size data.

[0038] As a further improvement of an embodiment of the present invention, the "reconstructing the portion representing the oral feature data on the oral three-dimensional model data to obtain the oral structure information" specifically includes: in the intraoral image data, determining at least two intraoral tissue feature points according to the oral feature data; based on the intraoral tissue feature points, at the feature missing position in the oral three-dimensional model, fitting tissue space feature points corresponding to the intraoral tissue feature points in accordance with the pixel size relationship, and forming the portion representing the oral feature data on the oral three-dimensional model data.

[0039] As a further improvement of an embodiment of the present invention, the intraoral tissue area includes a vestibule groove, the oral cavity feature data includes vestibule groove height data, and the intraoral tissue feature points include upper groove bottom feature points and lower groove bottom feature points.

[0040] As a further improvement of an embodiment of the present invention, the "fitting the tissue space feature points corresponding to the intraoral tissue feature points in accordance with the pixel size relationship" specifically includes: calculating the distance between the upper sulcus bottom feature point and the midpoint of the gingival margin of the corresponding tooth position to obtain the upper sulcus bottom spacing parameter, and calculating the corresponding upper sulcus bottom mapping parameter based on the upper sulcus bottom spacing parameter and the size mapping factor characterizing the pixel size relationship; fitting the upper space feature point characterizing the upper sulcus bottom position based on the reference feature coordinates of the target tooth position corresponding to the upper sulcus bottom feature point and the upper sulcus bottom mapping parameter; calculating the distance between the lower sulcus bottom feature point and the midpoint of the gingival margin of the corresponding tooth position to obtain the lower sulcus bottom spacing parameter, and calculating the corresponding lower sulcus bottom mapping parameter based on the lower sulcus bottom spacing parameter and the size mapping factor; fitting the lower space feature point characterizing the lower sulcus bottom position based on the reference feature coordinates of the target tooth position corresponding to the lower sulcus bottom feature point and the lower sulcus bottom mapping parameter.

[0041] As a further improvement of one embodiment of the present invention, the oral tissue area includes the vestibule groove and / or the root ridge, the oral feature data includes dental arch width data, and the oral tissue feature points include left groove bottom feature points and right groove bottom feature points.

[0042] As a further improvement of an embodiment of the present invention, the "determining at least two intraoral tissue feature points in the intraoral image data according to the oral feature data" specifically includes: determining all intraoral tissue feature points in the intraoral image according to the oral feature data; the "forming the part of the oral three-dimensional model data representing the oral feature data" specifically includes: fitting a tissue space distribution curve or a tissue space distribution surface according to the tissue space feature points, and using this as the part of the oral three-dimensional model data representing the oral feature data.

[0043] To achieve one of the above-mentioned purposes of the invention, an embodiment of the present invention provides a method for generating oral structure information, comprising: acquiring oral three-dimensional model data and corresponding intraoral image data; extracting oral feature data corresponding to the target intraoral tissue area in the oral three-dimensional model from the intraoral image data, establishing a pixel size relationship between the intraoral image data and the oral three-dimensional model data, and reconstructing the part representing the oral feature data on the oral three-dimensional model data to obtain the oral structure information.

[0044] To achieve one of the above-mentioned purposes of the invention, one embodiment of the present invention provides an oral structure information generation system, comprising a processor, a memory and a communication bus, wherein the processor and the memory communicate with each other via the communication bus; the memory is used to store application programs; and the processor is used to implement the steps of the oral structure information generation method described in any of the above-mentioned technical solutions when executing the application programs stored in the memory.

[0045] To achieve one of the above-mentioned purposes of the invention, an embodiment of the present invention provides a storage medium having an application stored thereon, and when the application is executed, the steps of the method for generating oral structure information described in any of the above-mentioned technical solutions are implemented.

[0046] To achieve one of the above-mentioned objects of the invention, an embodiment of the present invention provides an oral appliance, wherein the oral appliance is constructed according to oral structure information, and the oral structure information is generated according to the oral structure information generation method described in any of the above-mentioned technical solutions.

[0047] As a further improvement of one embodiment of the present invention, the oral appliance is used for training orofacial muscle function and / or for treating mouth breathing.

[0048] Compared with the prior art, the oral structure information generation method provided by the present invention analyzes the relative position relationship between the tooth position features and the model boundary features, and when it is determined that the oral 3D model does not meet the preset conditions, calls the corresponding intraoral image data to complete the features, thereby obtaining complete oral structure information based on the oral 3D model data, thereby reducing the requirements for extracting the oral 3D model, and simplifies the steps and operation logic on the basis of obtaining high-precision and complete and comprehensive oral structure features; the composite process between the intraoral image data and the oral 3D model data mainly reconstructs the features on the oral 3D model according to the established pixel size relationship, and the output oral structure information is based on the 3D data, has strong intuitiveness and accuracy, is convenient for medical workers to conduct further analysis, and is also convenient for patients or other relevant personnel to consult. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a schematic structural diagram of the oral three-dimensional model when the oral appliance is not installed in one embodiment of the present invention.

[0050] Figure 2 It is a schematic diagram of the structure of the three-dimensional model of the oral cavity when the oral appliance is installed in one embodiment of the present invention.

[0051] Figure 3 It is a schematic diagram of the structure of an oral structure information generation system in one embodiment of the present invention.

[0052] Figure 4 It is a schematic diagram of the steps of a method for generating oral structure information in one embodiment of the present invention.

[0053] Figure 5 It is a schematic diagram of the steps of a method for generating oral structure information in another embodiment of the present invention.

[0054] Figure 6 It is a first structural schematic diagram of another oral three-dimensional model in another embodiment of the present invention under a forward perspective.

[0055] Figure 7 It is a schematic diagram of the steps of a first embodiment of a method for generating oral structure information in another embodiment of the present invention.

[0056] Figure 8 It is a schematic diagram of some steps of a specific example of the first embodiment of the method for generating oral structure information in another embodiment of the present invention.

[0057] Fig. 9 It is a partial step diagram of another specific example of the first embodiment of the method for generating oral structure information in another embodiment of the present invention.

[0058] Fig.10 In another embodiment of the present invention, when executing the first embodiment of the oral structure information generation method, Figure 6 A partial enlarged schematic diagram of the Z1 part.

[0059] Fig.11 It is a schematic diagram of some steps of a second embodiment of a method for generating oral structure information in another embodiment of the present invention.

[0060] Fig.12 It is a structural schematic diagram of another oral three-dimensional model under a lateral perspective when executing the second embodiment of the oral structure information generation method in another embodiment of the present invention.

[0061] Fig.13 It is a structural schematic diagram of another oral three-dimensional model in a forward perspective when executing the second embodiment of the oral structure information generation method in another embodiment of the present invention.

[0062] Fig.14 It is a partial step schematic diagram of a third embodiment of a method for generating oral structure information in another embodiment of the present invention.

[0063] Fig.15 It is a structural schematic diagram of another oral three-dimensional model in a forward perspective when executing the third embodiment of the oral structure information generation method in another embodiment of the present invention.

[0064] Fig.16It is a schematic diagram of some steps of a fourth embodiment of a method for generating oral structure information in another embodiment of the present invention.

[0065] Fig.17 It is a schematic diagram of the state of distribution of distance training data on three-dimensional training model data when executing the fourth embodiment of the method for generating oral structure information in another embodiment of the present invention.

[0066] Fig.18 It is a schematic diagram of some steps of the fifth embodiment of the method for generating oral structure information in another embodiment of the present invention.

[0067] Fig.19 It is a schematic diagram of the steps of a method for generating oral structure information in another embodiment of the present invention.

[0068] Fig. 20 It is a schematic diagram of some steps of a first embodiment of a method for generating oral structure information in yet another embodiment of the present invention.

[0069] Fig.21 It is a second structural schematic diagram of another oral three-dimensional model in a forward perspective when executing the first embodiment of the oral structure information generation method in another embodiment of the present invention.

[0070] Fig. 22 It is a schematic diagram of some steps of a second embodiment of a method for generating oral structure information in yet another embodiment of the present invention.

[0071] Fig.23 It is a schematic diagram of some steps of a first embodiment of a method for generating oral structure information in one embodiment of the present invention.

[0072] Fig.24 It is a schematic diagram of some steps of a second embodiment of a method for generating oral structure information in one embodiment of the present invention.

[0073] Fig.25 is a method for generating oral structure information in accordance with an embodiment of the present invention. Fig.21 Schematic diagram of the partial enlargement of part Z2 in the middle and part Z3 in the corresponding intraoral image.

[0074] Fig.26 It is a schematic diagram of some steps of a first specific example of a second embodiment of a method for generating oral structure information in one embodiment of the present invention.

[0075] Fig. 27 It is a partial step diagram of a second specific example of a second embodiment of a method for generating oral structure information in one embodiment of the present invention.

[0076] Fig.28It is a schematic diagram of some steps of the third embodiment of the method for generating oral structure information in one embodiment of the present invention.

[0077] Fig.29 It is a schematic diagram of the steps of a method for generating oral structure information in another embodiment of the present invention. DETAILED DESCRIPTION

[0078] The present invention will be described in detail below in conjunction with the specific embodiments shown in the accompanying drawings. However, these embodiments do not limit the present invention, and any structural, methodological, or functional changes made by a person skilled in the art based on these embodiments are all within the scope of protection of the present invention.

[0079] It should be noted that the term "comprises" or any other variation thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device that includes a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In addition, the terms "first", "second", "third", "fourth", "fifth", etc. are used for descriptive purposes only and are not to be understood as indicating or implying relative importance.

[0080] The main idea of ​​the present invention is that after obtaining the oral three-dimensional model or its data, in order to avoid incomplete content of the oral three-dimensional model, such as missing features corresponding to the target tooth position or other areas, the intraoral photo image or its data corresponding to the oral three-dimensional model is called, on the one hand, a corresponding relationship between the intraoral photo image data and the oral three-dimensional model data is established, and on the other hand, the required information in the intraoral photo image data is extracted, and then the oral three-dimensional model data is supplemented or expanded, reducing the actual operation requirements for extracting the oral three-dimensional model, and outputting more intuitive model data for medical workers to analyze and diagnose. Preferably, before calling the intraoral photo image data, a step of judging the integrity of the oral three-dimensional model can also be included, thereby streamlining the process of generating oral structure information, so that the step of "generating oral structure information according to the oral three-dimensional model" can be automatically realized.

[0081] The following will further describe the various embodiments, technical principles and corresponding technical effects of the present invention in conjunction with the accompanying drawings. In one embodiment of the present invention, an oral appliance is provided. Figure 1 The installation and matching environment of the oral appliance or the oral appliance model is shown. Figure 1 The content shown in the figure can be interpreted as the actual internal environment of the human oral cavity, or as an extracted oral solid model or a modeled oral three-dimensional model. Figure 1 The structure shown takes the oral three-dimensional model 100 as an example. The left figure shows the rendered three-dimensional structure, and the right figure shows the contour structure corresponding to at least part of the structure of the oral three-dimensional model 100.

[0082] In a specific example of the present invention, the oral three-dimensional model 100 specifically includes teeth 11, labial and buccal mucosa 12, vestibule groove 13, root ridge 14 and labial frenulum 15. Among them, the vestibule groove 13 is also called the labial and buccal gingival groove, which can be interpreted as the upper and lower boundaries of the oral cavity. The vestibule groove 13 is iron-shoe-shaped as a whole, and is a groove-shaped tissue structure produced by the labial and buccal mucosa 12 migrating to the alveolar mucosa. Specifically, when the human head is in an upright state, the tooth positions in the oral three-dimensional model 100 are arranged in a conventional order. At this time, according to the FDI (Fédération Dentaire Internationale, World Dental Federation) tooth position representation method, tooth No. 11 is located to the upper left of tooth No. 31, and tooth No. 41 is located to the lower left of tooth No. 21. Based on this, the vestibule groove 13 can be defined as including an upper vestibule groove 131 and a lower vestibule groove 132, wherein the vestibule groove 13 located away from teeth 41 and 31 relative to teeth 11 and 21 is the upper vestibule groove 131, and the vestibule groove 13 located away from teeth 11 and 21 relative to teeth 41 and 31 is the lower vestibule groove 132. Of course, when observing the oral cavity three-dimensional model 100 from different perspectives, or setting the oral cavity three-dimensional model 100 to different positions and postures, the definitions of the upper and lower orientations, the upper vestibule groove 131, and the lower vestibule groove 132 can be adjusted accordingly, which is understood by those skilled in the art and will not be repeated here.

[0083] like Figure 1 and Figure 2 As shown, the oral appliance 200 provided by the present invention can be a physical oral appliance or a corresponding three-dimensional model or physical model. The oral appliance 200 is specifically customized and constructed according to the oral structure information, so that when the oral appliance 200 is matched or installed with the oral three-dimensional model 100 or the corresponding actual human oral environment, it can fit the oral tissue such as at least one of the above-mentioned teeth 11, labial and buccal mucosa 12, vestibule groove 13, root ridge 14 and lip frenulum 15 as much as possible, and improve the comfort or matching degree under the premise of realizing the function of the oral appliance 200 itself.

[0084] The oral appliance 200 can be specifically configured into multiple types. For example, the oral appliance 200 can be a dento-maxillary deformity appliance or retainer, or a device for training orofacial muscle function (also known as orofacial myofunctional therapy, OMT) and / or for treating mouth breathing. Corresponding to different types of oral appliances 200, the oral structure information on which the construction process is based may also be different. In the first case, the oral structure information includes the morphological information of all tissues inside the oral cavity, and specifically can include the morphological characteristics of the dentition crown, the morphological distribution of the root ridge 14, and the morphological distribution of the vestibule groove 13; in the second case, the oral appliance 200 is a dento-maxillary deformity appliance or retainer, then the oral structure information includes at least part of the morphological characteristics of the dentition crown, and preferably includes the morphological distribution of the root ridge 14; in the third case, the oral appliance 200 is an orofacial muscle barrier, then the oral structure information includes at least part of the morphological characteristics of the vestibule groove, and preferably includes the morphological characteristics of the dentition crown.

[0085] Regardless of which application scenario the oral appliance 200 belongs to, the size design of at least one dimension of the oral appliance 200 should be of a higher order of magnitude than that of the oral three-dimensional model 100, so that the oral appliance 200 can at least form a certain clearance distance with the root ridge 14 to prevent excessive squeezing of the gums, resulting in damage to the oral model or reduced wearing comfort. Based on this, in one embodiment, the oral three-dimensional model 100 includes a right distal root ridge 141 and a left distal root ridge 142, and the oral appliance 200 includes a right end 21 corresponding to the right distal root ridge 141, and a left end 22 corresponding to the left distal root ridge 142. In an embodiment in which the oral appliance 200 is configured for training orofacial muscle function and / or for treating mouth breathing, the right end 21 and the left end 22 can be specifically the end on the cheek screen away from the lip stopper, or the end on the cheek screen away from the breathing hole.

[0086] Among them, the right side end portion 21 can be defined as the end portion of the oral appliance 200 located on the right side of the distal root protrusion 141 away from the soft palate, and the distance of the right side end portion 21 away from the right side distal root protrusion 141 relative to the soft palate can be defined as the "certain clearance distance"; the left side end portion 22 can be defined as the end portion of the oral appliance 200 located on the left side of the distal root protrusion 142 away from the soft palate, and the distance of the left side end portion 22 away from the left side distal root protrusion 142 relative to the soft palate can be defined as the "certain clearance distance".

[0087] The "certain clearance distance" is freely selected according to the specific type or functional role of the oral appliance 200. For example, when the oral appliance 200 is configured as a malocclusion appliance or retainer, the distance between the right end 21 and the left end 22 can be less than or equal to the distance between the right distal root ridge 141 and the left distal root ridge 142, thereby constraining the teeth in the corresponding position to cause them to produce corresponding displacement or remain in their original positions. For another example, when the oral appliance 200 is configured to be used for training orofacial muscle function and / or for treating mouth breathing, or is configured as other devices for forming a barrier in the mouth, the distance between the right end 21 and the left end 22 may be greater than the distance between the right distal root ridge 141 and the left distal root ridge 142, and preferably, the difference between the spacing between the right end 21 and the left end 22 and the spacing between the right root ridge 141 and the left distal root ridge 142 is greater than or equal to 3 mm, so as not to excessively interfere with soft tissues such as the gums at the root ridge 142, thereby affecting the wearing experience or causing wear of the oral model.

[0088] The left distal root ridge 142 and the right distal root ridge 141 are the root ridges at the farthest tooth position relative to the tooth midline in terms of orientation. For adults, they usually refer to the root ridges of the maxillary second molars or the root ridges of the mandibular second molars. For children, they usually refer to the root ridges of the maxillary second deciduous molars or the root ridges of the mandibular second deciduous molars. The root ridge 14 at any of the above tooth positions can be interpreted as the intraoral tissue that wraps around the tooth root and protrudes away from the soft palate relative to the labial surface of the crown, and can specifically be the gingival portion located outside the root canal and the alveolar bone portion wrapped by the gingival portion.

[0089] Furthermore, when the oral appliance 200 is installed on or cooperated with the oral three-dimensional model 100 or the actual oral environment of the human body, the upper end portion close to the maxillary side can be fitted with the upper vestibule groove 131, and the lower end portion close to the mandibular side can be fitted with the lower vestibule groove 132, that is, the distance between the upper end portion and the lower end portion of the oral appliance 200 can be equal to the distance between the upper vestibule groove 131 and the lower vestibule groove 132.

[0090] Regarding the above-mentioned size relationship, considering that when extracting the corresponding oral three-dimensional model 100 of the actual oral environment of the human body, the distance between the upper vestibule groove 131 and the lower vestibule groove 132 may be greater than the distance between the two vestibule grooves when the human body is in a normal living state due to stretching, the above-mentioned "equal to" relationship can also be "slightly less than". Of course, in order to improve the training effect of the orofacial muscles and / or the therapeutic effect of mouth breathing, of course, the above-mentioned "equal to" relationship can also be "slightly greater than". Preferably, the cross-sectional shape and the extended distribution curve of the upper end of the oral appliance 200 can also fit the distribution curve and tissue morphology of the upper vestibule groove 131, and the cross-sectional shape and the extended distribution curve of the lower end of the oral appliance 200 can also fit the distribution curve and tissue morphology of the lower vestibule groove 132, so that the oral appliance 200 as a whole is also configured as an iron hoof shape.

[0091] Under the iron-shoe-shaped overall configuration, in order to avoid the lip frenulum 15 in the mouth and prevent the oral appliance 200 from exerting pressure on it, the middle of the upper end and the middle of the lower end of the oral appliance 200 may be provided with avoidance portions recessed toward the geometric center of the oral appliance 200. At the same time, considering that the shape of the lip frenulum 15 may be different in different oral models and actual human oral environments, the width of the avoidance portion extending along the length extension direction of the oral appliance 200 should be at least greater than or equal to the width of the lip frenulum 15 on the three-dimensional oral model 100 to prevent unnecessary limitation of the soft tissue of the lip frenulum 15 and cause pain to the wearer.

[0092] The above-mentioned description of the characteristics of the oral appliance 200, on the one hand, can be used as a limitation of the morphological characteristics of the oral appliance 200 itself, so as to achieve the corresponding technical effects mentioned above; on the other hand, in one embodiment, the oral appliance 200 is configured to be constructed based on oral structure information, and the oral structure information is generated according to an oral structure information generation method. Based on this, the above-mentioned description of the oral appliance 200 can be interpreted as the beneficial effects of the oral structure information or the oral structure information generation method. In other words, when executing the steps of the oral structure information generation method provided by the present invention, a kind of oral structure information can be generated so that the corresponding oral appliance produced has any of the above-mentioned characteristics and technical solutions.

[0093] The present invention also provides a storage medium, which can be specifically a computer-readable storage medium. The storage medium can be set in a computer and store an application program. In this case, the storage medium can be any available medium that the computer can access data, or can be a storage device such as a server or data center that includes one or more available media. The available medium can be a magnetic medium such as a floppy disk, a hard disk, a magnetic tape, or an optical medium such as a DVD (Digital Video Disc), or a semiconductor medium such as an SSD (Solid State Disk). When the application program is executed, the steps of a method for generating oral structure information are implemented to at least perform: obtaining intraoral image data and oral three-dimensional model data, establishing pixel size relationships, and reconstructing oral feature data on the oral three-dimensional model data. Preferably, a step of judging the integrity of the oral three-dimensional model data is performed.

[0094] One embodiment of the present invention further provides a Figure 3 The oral structure information generation system 300 shown includes a processor 31, a memory 33 and a communication bus 34. The processor 31 and the memory 33 communicate with each other through the communication bus 34. In order to further expand the functions of the oral structure information generation system 300, the oral structure information generation system 300 may also include a communication interface 32 for the oral structure information generation system 300 to communicate with other systems or devices such as medical worker operating systems / equipment, patient clients, manufacturer / warehouse management systems, or manufacturing / warehouse management equipment. Similarly, the processor 31, the communication interface 32 and the memory 33 can communicate with each other through the communication bus 34.

[0095] Correspondingly, the memory 33 is used to store the application program; the processor 31 is used to execute the application program stored in the memory 33, and the application program may be an application program stored in the storage medium mentioned above, that is, the storage medium may be configured to be at least contained in the memory 33. Based on this, when executing the application program, the processor 31 may implement a method for generating oral structure information, which may specifically include the steps of obtaining intraoral image data and oral 3D model data, establishing pixel size relationship, and reconstructing oral feature data on the oral 3D model data. Preferably, it includes the step of judging the integrity of the oral 3D model data.

[0096] The communication bus 34 may be a PCI bus (Peripheral Component Interconnect) or an EISA bus (Extended Industry Standard Architecture). The communication bus 34 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0097] The memory 33 may include a RAM (Random Access Memory) or a NVM (Non-Volatile Memory), such as at least one disk memory. The processor 31 may be a general-purpose processor, including a CPU (Central Processing Unit), a NP (Network Processor), etc. It may also be a DSP (Digital Signal Processing), an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array) or other programmable logic devices, discrete gates or transistor logic devices, or discrete hardware components.

[0098] Of course, although the present invention provides an oral structure information generating system 300 as described above, it can be seen based on the description of the oral structure information generating system 300 that its internal components can also be integrated into one device after being matched with the embodiments. Based on this, the oral structure information generating system 300 does not only refer to large systems such as field bus control systems, but can also refer to small circuit systems or control systems in oral structure information generating devices.

[0099] like Figure 4 As shown, an embodiment of the present invention provides a method for generating oral structure information. The application or instruction corresponding to the method can be installed in the above storage medium and / or the above oral structure information generation system 300 to achieve the technical effect of generating oral structure information. The method for generating oral structure information can specifically include the following steps.

[0100] Step 40, obtaining oral three-dimensional model data, and determining whether the intraoral tissue region specified in the oral three-dimensional model meets a preset integrity condition.

[0101] The oral three-dimensional model data refers to data information that can establish an oral three-dimensional model. The source of the data information can be electronic data information such as multi-angle intraoral images, oral scanning models, or physical models such as oral silicone models and oral plaster models. For electronic data information, it can be formed through composite reorganization and restoration to form oral three-dimensional model data that is as comprehensive as possible (but not required to completely include all oral tissue features); for physical models, it can be reconstructed and occluded through CT (Computed Tomography). Preferably, the oral three-dimensional model can be a three-dimensional model that characterizes the maxillofacial structure characteristics of the patient in an occluded state.

[0102] The "designated oral tissue area" can be interpreted as an area on the oral three-dimensional model that is determined to adapt to different diagnosis and treatment purposes or task types. The standard for determination can be defined by medical workers, or it can be automatically determined by the task type preset by the system. For the former, it can specifically include the step of "obtaining the oral tissue selection instruction"; for the latter, it can specifically include the step of "determining the task type and the corresponding oral tissue area according to the analysis instruction". The basis for determining the oral tissue area can be based on the position of different teeth in the oral three-dimensional model, or it can be determined after segmentation based on indicators such as point cloud features and grayscale features of other oral tissues. The oral tissue area includes, but is not limited to, the vestibule sulcus, lip frenulum, dental arch and other parts.

[0103] The integrity condition may be a preset indicator requirement for measuring the overall integrity of the oral three-dimensional model. The judgment result may be manually identified to obtain an integrity score, which is then compared by the system. Based on this, step 40 may include: obtaining an integrity score based on the oral three-dimensional model data, and judging whether at least the intraoral tissue area in the oral three-dimensional model meets the preset integrity condition.

[0104] The judgment result of the integrity condition can also be obtained by first evaluating through artificial intelligence and machine learning means such as a pre-built neural network model, and then comparing. Based on this, step 40 can include: analyzing the oral three-dimensional model to obtain an integrity score corresponding to at least the intraoral tissue area, and judging whether at least the intraoral tissue area in the oral three-dimensional model meets the preset integrity condition. Wherein, the integrity score at least points to the intraoral tissue area.

[0105] The process of evaluating the integrity of the oral three-dimensional model includes, but is not limited to, performing positional relationship analysis, grayscale analysis, point cloud density analysis, and image or model continuity analysis on at least the intraoral tissue area.

[0106] Of course, the setting of the above integrity condition is not limited to the local integrity judgment of some important intraoral tissue areas, but can also be set for the entire oral 3D model. Correspondingly, the integrity evaluation process can also be performed on the entire oral 3D model.

[0107] If not satisfied, jump to step 43, obtain the corresponding intraoral image data, extract the oral feature data corresponding to the intraoral tissue area in the intraoral image data, establish the pixel size relationship between the intraoral image data and the oral three-dimensional model data, and reconstruct the part representing the oral feature data on the oral three-dimensional model data to obtain the oral structure information.

[0108] The intraoral image data corresponds to the oral three-dimensional model data, and this correspondence is not limited to the two having to point to the same object, but as long as they can represent roughly the same structural features. For example, the oral three-dimensional model data can be based on the actual oral environment of the human body and established based on an oral scanning model, while the intraoral image can be established based on an oral silicone model, an oral gypsum model, etc., as long as the oral silicone model or the oral gypsum model points to the same oral tissue structure as the oral scanning model. However, it should be noted that the intraoral image data needs to include the intraoral tissue area corresponding to the oral feature data to be reconstructed.

[0109] The establishment of the pixel size relationship is not limited to scaling the intraoral image and the oral 3D model accordingly to match the two, as long as the oral feature data can be mapped onto the oral 3D model or its data while retaining the relative position relationship.

[0110] In this way, the integrity of at least part of the area on the oral three-dimensional model can be determined, and at least part of the area on the oral three-dimensional model can be selectively reconstructed to complete the missing oral feature data or information on the oral three-dimensional model, thereby generating complete oral structure information for medical workers to analyze and diagnose.

[0111] like Figure 5 As shown, another embodiment of the present invention provides a method for generating oral structure information, specifically providing specific steps 41 and 42 for step 40, preferably determining whether the oral three-dimensional model meets the integrity condition by analyzing the target tooth position characteristics and the model boundary characteristics. It can be understood that the application or instruction corresponding to the method can also be carried in the above-mentioned storage medium and / or the above-mentioned oral structure information generation system 300. The oral structure information generation method can specifically include the following steps.

[0112] Step 41, obtain oral three-dimensional model data, and calculate the positional relationship between the target tooth position and the model boundary and / or the positional relationship between the maxillary model boundary and the mandibular model boundary based on the target tooth position features in the oral three-dimensional model data and the model boundary features corresponding to the target tooth position features.

[0113] The target tooth position feature has two meanings. First, the "target" means that the target tooth position feature actually points to the part that needs to be analyzed in the three-dimensional model of the oral cavity, and corresponds to which specific position needs to be analyzed. Regarding the steps of position analysis and determination, there are multiple technical solutions. In other words, there are multiple technical solutions for how to determine the target tooth position feature. In one technical solution, the position pointed to by the target tooth position feature can be defined by medical workers. For example, if it is necessary to check the caries condition of a specific tooth position, then the first molar tooth position can be selected as the target tooth position, and a certain feature of the tooth position can be further analyzed as the target tooth position feature; for example, if it is necessary to check the numerical size of the vestibule groove in a certain dimension, then the tooth position around the entire vestibule groove or the local periphery can be selected as the target tooth position, and then the target tooth position feature can be obtained. In another technical solution, the position pointed to by the target tooth position feature can be determined according to the target task type, which can be a task of making oral appliances such as orthodontic appliances or orofacial muscle barriers, or a task of checking the integrity of the dentition in the mouth (for example, whether there are missing teeth). For the former, it can specifically include the steps of: obtaining target appliance type information and determining the target oral tissue; and determining the corresponding tooth position as the target tooth position based on the target oral tissue.

[0114] Secondly, the “feature” indicates its functional role, that is, the target tooth position feature is actually a feature information that can characterize the relative position of the target tooth position, such as the coordinates of a feature point on the crown of the target tooth position, or the coordinates of a feature point on the root ridge of the target tooth position. The feature point and its coordinates can not only be the coordinates of a point on the edge or surface of an oral tissue on the corresponding tooth position, but also the coordinates of the geometric center point of an oral tissue on the tooth position, or the coordinates of the center point of the whole set formed by multiple oral tissues on the tooth position. Before forming the target tooth position feature, it can usually include a process of determining the target tooth position number (this process can also be reused when determining the target tooth position pointing to the model position), and thus can specifically include the steps of: traversing all tooth position data in the oral three-dimensional model data to determine the number of each tooth position; according to the number of each tooth position, determining a feature coordinate point on each tooth position as the target tooth position feature.

[0115] The model boundary feature, at the level of the meaning of "boundary", can be a boundary point or boundary distribution curve on the oral three-dimensional model. At the level of "model boundary" and its corresponding relationship with the target tooth position feature, it can be a point or coordinate located within a certain area or volume range of the target tooth position feature and at the boundary of the oral three-dimensional model; it can also be a model boundary point or coordinate obtained by traversing and searching according to the target tooth position feature in a preset direction, so that the corresponding relationship between the two can be used to reflect the positional relationship between the tooth position and the model boundary, or the positional relationship between the upper and lower jaw model boundaries, and further used to determine whether the oral three-dimensional model meets the integrity condition at least in the corresponding oral tissue area, and even use it to perform operations such as reconstruction of features on the oral three-dimensional model.

[0116] Combination Figure 6 As shown, the first tooth position 11a in the oral three-dimensional model 100 can be determined as the target tooth position, and its first gingival margin midpoint 11A can be determined as the coordinate point characterizing the target tooth position feature. When the implementation method of traversing the model boundary feature corresponding to the target tooth position feature along the preset direction is adopted, the first model boundary point 10A corresponding to the first gingival margin midpoint 11A can be retrieved on the oral three-dimensional model 100 along the first reference direction D11, and the model boundary feature can be characterized by the first model boundary point 10A. Of course, there can be multiple groups of the target tooth position, model boundary, target tooth position feature and model boundary feature. For example, if the second tooth position 11b is used as the target tooth position, the second gingival margin midpoint 11B and the second model boundary point 10B can be obtained in sequence based on the same scheme. For another example, if the fourth tooth position 11d is used as the target tooth position, the fourth gingival margin midpoint and the fourth model boundary point can be determined in sequence in the mandible of the oral three-dimensional model 100 based on the same scheme. Based on this, the positional relationship between the maxillary model boundary and the mandibular model boundary in step 41 can be calculated using the first model boundary point 10A or the second model boundary point 10B and the fourth model boundary point. Here, it can be understood that the first reference direction D11 can of course include the first reference reverse direction D11' corresponding to it, and the two directions can be defined as one direction, and adaptively, when retrieving the maxillary model boundary features, the first reference direction D11 is used to search away from the occlusal plane (the imaginary plane formed by the mesial adjacent point of the maxillary central incisor to the mesio-buccal cusps of the bilateral first molars), and when retrieving the mandibular model boundary features, the first reference reverse direction D11' is used to search away from the occlusal plane.

[0117] The oral three-dimensional model 100 includes a first model boundary 10a corresponding to the first model boundary point 10A, and a second model boundary 10b corresponding to the second model boundary point 10B. The first model boundary 10a and the second model boundary 10b can be interpreted as model boundaries formed by fitting the model boundary points obtained by traversing and searching for multiple tooth position feature coordinate points, or can be interpreted as model boundaries that can intersect with reference lines corresponding to multiple tooth position feature coordinate points to form corresponding model boundary points. The former enables it to be known on the user side, and the latter embodies its essence that actually exists in the oral three-dimensional model 100.

[0118] Correspondingly, the first tooth position 11a in the oral three-dimensional model 100 can also be determined as the target tooth position, but the coordinate point characterizing the target tooth position feature can be determined based on the geometric center of the area enclosed by the convex hull contour of the crown of the first tooth position 11a, that is, the center point 11A' of the labial surface of the first crown can be used to characterize the target tooth position feature. At the same time, a second reference direction D12 that is distinguished from the first reference direction D11 can also be used to traverse and search at an inclination angle relative to the extension direction of the tooth length, thereby retrieving the first reference boundary point 10A' corresponding to the center point 11A' of the labial surface of the first crown to characterize the model boundary feature. The second reference direction D12 can be the direction used by all tooth positions to determine the model boundary feature. Of course, the directions used by each tooth position to determine the model boundary feature can also be distinguished from each other.

[0119] In one embodiment, the actual direction reflected by the second reference direction D12 on each tooth position may be different from each other, thereby being able to generate derived directions such as the third reference direction D13. The determination of the second reference direction D12 and the third reference direction D13 may include the steps of: determining the global reference center point C1 of the oral three-dimensional model on the side of the crown labial surface; using the global reference center point C1 and the first crown labial surface center point 11A' as reference lines, and using the extension direction of the reference line as the second reference direction D12 corresponding to the first tooth position 11a. Based on this, corresponding to the second tooth position 11b, the global reference center point C1 and the corresponding second crown labial surface center point can also be used as reference lines to determine the third reference direction D13. Among them, the global reference center point C1 can preferably be the midpoint of the line segment formed by at least part of the tooth center line, and specifically can be the midpoint of the projection line segment of the tooth center line on the oral three-dimensional model 100.

[0120] In another embodiment, the direction of traversing and retrieving the model boundary features can also be determined by determining a combination of local center points and edge points. Preferably, the local center point can be the center point of the labial surface of the crown, and the edge point can be the midpoint of the gingival margin. For example, the third tooth position 11c in the oral three-dimensional model 100 can be determined as the target tooth position, and the third crown labial surface center point C2 corresponding to the third tooth position 11c is determined as the local center point according to the method steps provided in the previous text, and the third gingival margin midpoint 11C corresponding to the third tooth position 11c is determined as the edge point according to the method steps provided in the previous text. Then, the third crown labial surface center point C2 and the third gingival margin midpoint 11C are used as reference lines, and the extension direction of the reference line is used as the fourth reference direction D14 corresponding to the third tooth position 11c. In this way, the third model boundary point 10C corresponding to the third tooth position 11c can be obtained.

[0121] At this point, those skilled in the art can understand that the present invention includes at least multiple implementation methods such as the midpoint of the gingival margin and the center point of the labial surface of the crown in determining the characteristics of the target tooth position; at least multiple implementation methods such as determining by area and determining by direction in determining the boundary characteristics of the model; in the implementation method of determining the boundary characteristics of the model by direction, at least multiple implementation methods such as determining by fixed direction, determining by the center point of the labial surface of the crown and the global reference center point, and determining by the midpoint of the gingival margin and the local reference center point are included. In addition, the implementation methods that can be simply deduced from the above-mentioned multiple implementation methods and embodiments are all included in the scope of the technical solution provided by the present invention.

[0122] It is worth noting that in step 41, the calculation part includes three technical solutions, one is to calculate the positional relationship between the target tooth position and the model boundary, the second is to calculate the positional relationship between the maxillary model boundary and the mandibular model boundary, and the third is to calculate the positional relationship between the target tooth position and the model boundary, and calculate the positional relationship between the maxillary model boundary and the mandibular model boundary.

[0123] Step 42, comparing the positional relationship with a preset position condition corresponding to the positional relationship, and determining whether the intraoral tissue region corresponding to the target tooth position in the oral three-dimensional model satisfies the preset integrity condition.

[0124] Combination Figure 6As shown, the preset position condition can have various forms, such as "up", "down", "left", and "right" indicating the orientation, and intervals and thresholds indicating the relationship between numerical values. For the former, it can be judged according to a single feature, for example, the position of the model boundary feature coordinate point retrieved on the oral three-dimensional model 100 is used for judgment. Under the premise that the intraoral tissue area points to the height of the vestibule groove, for the maxilla, it can be set that when the model boundary feature coordinate point is below the set standard coordinate point, it is judged that at least the preset position condition is not met, and the integrity condition may not be met; under the premise that the intraoral tissue area points to the width of the vestibule groove, for the left side of the oral cavity (the side where the first tooth position 11a and the third tooth position 11c are located), it can be set that when the model boundary feature coordinate point is to the left of the set standard coordinate point, it is judged that at least the preset position condition is not met, and the integrity condition may not be met. It can be seen that the setting of the preset position condition can be related to the index of the intraoral tissue area pointing to, and can also be related to the position of the target tooth position in the oral three-dimensional model 100. In addition, judgment can also be made according to the geometric figures or distribution ranges jointly formed by the model boundary feature coordinate points and the target tooth position feature coordinate points, for example, whether the line connecting the model boundary feature coordinate points and the target tooth position feature coordinate points intersects with a preset line segment or area, or for another example, a circle is made with the line connecting the model boundary feature coordinate points and the target tooth position feature coordinate points as the diameter, and it is judged whether the area range of the circle falls within a preset area range or overlaps with a preset line segment.

[0125] For the latter, it is preferred to use the Euclidean distance between the model boundary feature coordinate point and the target tooth position feature coordinate point, or the length of the projection line segment of the connecting line in a certain direction as the analysis object, and compare the numerical value of the analysis object with the set threshold or range to determine whether the intraoral tissue area corresponding to the target tooth position meets the preset position condition. The integrity condition can preferably be a derivative condition for whether the preset position condition is met or not. For example, a threshold is set for the number of parts that do not meet the preset position condition. After the number of intraoral tissue parts that do not meet the preset position condition reaches the threshold, it is determined that the oral three-dimensional model cannot be put into use directly, or it is meaningless to continue analyzing the oral three-dimensional model.

[0126] Based on this, the present invention provides a variety of implementation methods for calculating the "positional relationship", and the integrity condition and the positional condition are at least indicators corresponding to the positional relationship. For example, in a preferred embodiment, taking the first tooth position 11a as an example, if the parameter used to characterize the positional relationship is the length of the projection line segment of the first gingival margin midpoint 11A and the first model boundary point 10A in the first reference direction D11, then the corresponding preset positional condition should also be the length of the projection line segment of the gingival margin midpoint of the central incisor (which can be limited to the left maxillary central incisor when greater accuracy is required) corresponding to the first tooth position 11a and a certain intraoral tissue feature point in its first reference direction D11 in the first reference direction D11. In summary, the preset positional condition can at least be consistent with the positional relationship in terms of the selection of direction, tooth position, and point. The integrity condition is preferably a judgment condition formed by integrating all judgment results relative to the preset positional condition, thereby establishing an indirect correlation with the positional relationship.

[0127] If not satisfied, jump to step 43, obtain the corresponding intraoral image data, extract the oral feature data corresponding to the intraoral tissue area in the intraoral image data, establish the pixel size relationship between the intraoral image data and the oral three-dimensional model data, and reconstruct the part representing the oral feature data on the oral three-dimensional model data to obtain the oral structure information.

[0128] When it is determined through steps 41 to 42 that the vestibule groove feature above the maxillary central incisor is missing and does not meet the preset integrity condition, the acquired intraoral image data should at least include the vestibule groove tissue part above the maxillary central incisor. It can be seen that the contribution of step 42 to step 43 is not only "whether the integrity condition is met", but also the specific basis and relevant circumstances for making such integrity condition judgment results, which at least may include the position of the part or tooth position that does not meet the preset position condition in the oral cavity.

[0129] Combination Figure 6 To illustrate, referring to the first tooth position 11a, when traversing and searching along the first reference direction D11, if it is determined that the first gingival margin midpoint 11A and the first model boundary point 10A do not meet the preset position condition (for example, the preset position of the first model boundary point 10A should be above point 13A and along the first reference direction D11), the oral feature data corresponding to the vestibule groove is analyzed through the intraoral image data to be the first vestibule groove bottom coordinate point 13A, or the distance length between the first vestibule groove bottom coordinate point 13A and the first gingival margin midpoint 11A. At this time, it is necessary to reconstruct at least the first vestibule groove bottom coordinate point 14A on the oral three-dimensional model 100 according to the position of the first vestibule groove bottom coordinate point 13A in the intraoral image data and the pixel size relationship, for example, to form Figure 6The “dot” pointed to by the first vestibule groove bottom coordinate point 13A in the figure. Obviously, in this process, it is not required to enlarge the intraoral image to the same size as the oral three-dimensional model 100. In addition, based on the same or similar steps, the first vestibule groove bottom reference point 13A′, the second vestibule groove bottom coordinate point 13B and the third vestibule groove bottom coordinate point 13C can be formed accordingly.

[0130] Of course, this does not mean that the present invention excludes the technical solution of unifying the size of the intraoral image and the oral three-dimensional model 100, especially when it is necessary to fit the curved surface (such as Figure 6 The first tissue spatial distribution surface Sa shown, or the third tissue spatial distribution surface at the third tooth position 11c) or the curve (such as Figure 6 When the first tissue spatial distribution curve 13a or the second tissue spatial distribution curve 13b is used, a pixel size relationship can also be established for multiple oral feature data to achieve an effect similar to scaling the intraoral tissue area.

[0131] like Figure 7 As shown, the present invention provides a first embodiment of a method for generating oral structure information based on the above-mentioned implementation mode, and the first embodiment specifically includes the following steps.

[0132] Step 410, obtaining oral cavity three-dimensional model data.

[0133] Step 411, determining at least one tooth position on the oral three-dimensional model as a target tooth position, and calculating the reference feature coordinates of the target tooth position based on a preset feature recognition rule to characterize the target tooth position features.

[0134] The reference feature coordinates may be the coordinates of the midpoint of the gingival margin or the coordinates of the center point of the labial surface of the crown, etc., so as to at least characterize the positional features of the target tooth. Of course, the coordinates of multiple points distributed on the gingival margin and the coordinates of multiple points distributed on the labial surface of the crown can also be used to characterize the morphological features of the target tooth; the positional features of the target tooth can also be characterized by using the midpoint coordinates of the incisor ridge of the incisor, the cusp coordinates of the canine, or the center point coordinates of the occlusal surface of the molar as the reference feature coordinates.

[0135] The preset feature recognition rule can be to determine the contour features of the crown of the target tooth through the spatial relative position of the pixel gray value or the point cloud set, and find the coordinates of a spatial point farthest from the occlusal plane based on the spatial points on the convex hull contour as the reference feature coordinates that can reflect the coordinates of the midpoint of the gingival margin. Of course, when the reference feature coordinates point to other positions of the target tooth, the extraction process can also have other derived technical solutions.

[0136] Step 412: According to the reference feature coordinates, boundary feature coordinates corresponding to the reference feature coordinates on the oral cavity three-dimensional model are determined along the first direction to characterize the model boundary features.

[0137] Combination Figure 6 As shown, the first direction can be correspondingly defined as D1. When different schemes for determining directions are adopted, the first direction D1 can point to any of the first reference direction D11, the first reference opposite direction D11', the second reference direction D12, the third reference direction D13 or the fourth reference direction D14, or can point to other directions that are not explicitly stated in the foregoing text but that are within the capabilities of those skilled in the art to make.

[0138] The boundary feature coordinates can be the above-mentioned model boundary points or model boundary coordinate points. In this way, feature points corresponding to the target tooth position features and representing the model boundary features can be obtained through traversal retrieval, and the reference feature coordinates and boundary feature coordinates can be used to form the relative position relationship between the target tooth position and the model boundary, and the abstract position relationship can be quantified into a calculable and clear numerical value, which is convenient for subsequent use in model integrity judgment.

[0139] Step 413, calculate and obtain the characteristic distance value between the reference characteristic coordinates and the boundary characteristic coordinates to characterize the positional relationship between the target tooth position and the model boundary.

[0140] The characteristic distance value can be the Euclidean distance between the reference characteristic coordinates and the boundary characteristic coordinates, or it can be the Euclidean distance or the projection of the line connecting two points in a certain direction. The present invention is not limited to this. As mentioned above, it is only necessary to set the calculation method of the corresponding preset position conditions to be consistent with the calculation method of the characteristic distance value that characterizes the position relationship to achieve the expected technical effect. Quantifying the position relationship between the target tooth position and the model boundary using the characteristic distance value can facilitate the setting of the preset position conditions and the calculation in the subsequent judgment process. Compared with fitting the position relationship using other spatial indicators such as volume or area, it can avoid the reduction of the credibility of the data in the transformation and mapping process due to the dimensional difference between the intraoral image and the three-dimensional model of the oral cavity.

[0141] Step 421 , compare the characteristic distance value with the distance integrity criterion value representing the preset position condition.

[0142] In other words, the technical solution provided by the present invention sets a distance integrity criterion value to characterize the preset position condition. This point can be interpreted as the distance integrity criterion value is part of the preset position condition, and the present invention does not exclude setting multiple dimensions of integrity criterion values ​​under the preset position condition, so as to comprehensively judge from multiple dimensions whether the positional relationship between the target tooth position and the model boundary satisfies the preset positional relationship. Among them, the integrity criterion value can also be replaced and selected according to the type of target tooth position and target oral tissue. For example, when judging the maxillofacial protrusion amplitude (corresponding to the maxillary incisor tooth position and the mandibular incisor tooth position) or the dental arch curvature (corresponding to all maxillary tooth positions or all mandibular tooth positions), an arc integrity criterion value is set.

[0143] Step 422, judging whether the intraoral tissue region corresponding to the target tooth position in the oral three-dimensional model satisfies a preset integrity condition based on the comparison result.

[0144] The correlation between the integrity condition and the comparison result is that when it is only necessary to determine whether the intraoral tissue area corresponding to a single target tooth position is complete, the comparison result can be directly used as a sufficient condition for determining whether the integrity condition is met; when it is necessary to perform integrity judgments on the intraoral tissue areas corresponding to multiple target tooth positions using characteristic distance values ​​respectively, it is necessary to set an integrity condition for all comparison results, so as to measure whether the overall intraoral tissue area corresponding to multiple target tooth positions meets the expected integrity requirements.

[0145] Combination Figure 6 As shown, the maxillary left central incisor tooth position, the maxillary left lateral incisor tooth position, and the maxillary left canine tooth position are used as the target tooth positions, and the comparison results corresponding to the three tooth positions may not meet the preset position conditions. At this time, it can be determined that the oral tissue areas corresponding to the above three tooth positions are incomplete. The maxillary right central incisor tooth position, the maxillary right lateral incisor tooth position, and the maxillary right canine tooth position are used as the target tooth positions, and only the comparison result of the maxillary right lateral incisor tooth position may show that it does not meet the preset position conditions. At this time, if the preset integrity conditions are set to be relatively loose, the oral tissue areas corresponding to the above three tooth positions can be considered to meet the integrity conditions as a whole. It can be understood that the above setting of the preset integrity conditions focuses on judging the number of teeth that do not meet the preset position conditions. Of course, it can also be based on the difference between the sum of the characteristic distances and the integrity criterion value, such as calculating its mean, variance and other indicators, to measure whether it meets the preset integrity conditions.

[0146] If not satisfied, jump to step 43, obtain the corresponding intraoral image data, extract the oral feature data corresponding to the intraoral tissue area in the intraoral image data, establish the pixel size relationship between the intraoral image data and the oral three-dimensional model data, and reconstruct the part representing the oral feature data on the oral three-dimensional model data to obtain the oral structure information.

[0147] The first embodiment provided by the present invention can simplify the expression of positional relationships, concretize abstract content into the distance between two coordinates, and can judge the integrity of the oral three-dimensional model based on a simple coordinate relationship, thereby realizing the judgment of complex abstract content on the basis of streamlining the algorithm logic.

[0148] Preferably, the "calculating the reference feature coordinates of the target tooth position based on preset feature recognition rules" specifically includes: determining and using the coordinates of the midpoint of the gingival margin of the target tooth position as the reference feature coordinates of the target tooth position. Since the midpoint of the gingival margin is located near the position on the crown of the target tooth position that is farthest from the occlusal plane, it is relatively closer to other intraoral tissues on the three-dimensional model of the oral cavity. Using the coordinates of this point as the reference feature coordinates can reduce the number of spatial feature points traversed on the three-dimensional model of the oral cavity and speed up the calculation.

[0149] In a specific example of the first embodiment, the reference feature coordinates are located in a preset model coordinate system, and the origin of the model coordinate system is located in the occlusal plane. Figure 7 and Figure 8 As shown, the step 411 may specifically include the following steps. It is worth noting that the specific example includes steps 410 to 43, but they will not be described in detail below.

[0150] Step 51, determining all spatial points on the target tooth crown located on the labial side to form a set of spatial points.

[0151] The external effect of the oral three-dimensional model is a three-dimensional figure composed of multiple faces, but its essence is still formed by computer software through fitting multiple spatial points. Based on this, the oral three-dimensional model can be analyzed by setting a certain density to obtain point cloud data or point set data distributed on the oral three-dimensional model, that is, the spatial point set.

[0152] It should be noted here that the "determine all spatial points on the crown of the target tooth position located on the labial side" can be interpreted as at least being able to confirm the above-mentioned spatial points on the crown of the target tooth position. On the one hand, step 51 can be to analyze the point cloud data of the entire oral three-dimensional model according to a preset density, and execute the following steps 52 to 56 to extract and analyze the set of spatial points corresponding to the target tooth position; on the other hand, before step 51, a neural network algorithm can be called or other methods can be used to select an area of ​​interest on the oral three-dimensional model, such as the area where the target tooth position or the entire dentition is located, which can also be analyzed to obtain the set of spatial points and reduce the complexity of the algorithm.

[0153] The reason for selecting all the spatial points on the lip side depends on the demand for oral feature data. In this embodiment, the main focus is on the characteristics of the intraoral tissues located on the lip side of the dentition, such as the vestibule groove, maxillofacial protrusion amplitude and lip frenulum. Among them, the above-mentioned lip surface can also be interpreted as the buccal surface for molars. Of course, in other embodiments, such as when it is necessary to pay attention to the development of the alveolar bone or the characteristics of the intraoral tissues such as the hard palate and soft palate, all the spatial points on the lingual or palatal side of the crown of the target tooth position can be calculated based on the preset feature recognition rules to form the set of spatial points.

[0154] Step 52, select the spatial point with the smallest coordinate value in the spatial point set as the pole to establish a polar coordinate system, and arrange the other spatial points in ascending order according to the polar angle and polar diameter to form a feature traversal sequence.

[0155] Combination Figure 6 and Fig.10 As shown, the above technical solutions with reference feature coordinates and boundary feature coordinates as operation objects are all established under the premise that the above coordinates are in a preset model coordinate system. The model coordinate system can be a number axis, a plane rectangular coordinate system or a space rectangular coordinate system, etc. Preferably, the model rectangular coordinate system includes at least two coordinate axes, so as to reflect the relative position relationship between the reference feature coordinates and the boundary feature coordinates. In order to form the mutual symmetry of the maxillary and mandibular, preferably the mutual symmetry of the left side of the dentition and the right side of the dentition, the origin of the model coordinate system can be set in the occlusal plane, and preferably at the intersection of the occlusal plane and the tooth midline. Of course, in other embodiments, the origin can also be set at the above-mentioned global reference center point C1. Based on this, the "spatial point with the smallest coordinate value" represents the spatial point at least closest to the occlusal plane among all the spatial points located on the labial side of the crown of the target tooth position. At this time, we can consider that the spatial point is on the edge of the crown of the target tooth position, that is, it can be considered as a point located on the contour of the crown of the target tooth position.

[0156] For example, the model coordinate system may be a plane rectangular coordinate system including the first coordinate axis Rx and the third coordinate axis Rz, or a space rectangular coordinate system including the first coordinate axis Rx, the third coordinate axis Rz and the second coordinate axis Ry, with the origin Ro located in the occlusal plane ( Figure 6 In order to avoid interference and unclear drawing, it is placed outside the oral 3D model. Figure 6 The Z1 part is formed Fig.10 From the locally enlarged schematic diagram, it can be seen that when the first tooth position 11a is selected as the target tooth position, the origin Ro is on the side of the incisal end of the first tooth position 11a, and when the first tooth position 11a is the maxillary left central incisor, and the origin Ro is preferably the intersection of the occlusal plane and the tooth midline, the origin Ro is located on the mesial surface side of the first tooth position 11a. Based on this, a model coordinate system can be established in which the first coordinate axis Rx extends along the width direction of the first tooth position 11a, and the third coordinate axis Rz extends along the length direction of the first tooth position 11a (or the extension direction of the long axis of the tooth, or specifically for the central incisor, along the tooth midline direction).

[0157] The “minimum coordinate value” may refer to the minimum coordinate value in the direction of the first coordinate axis Rx, the minimum coordinate value in the direction of the third coordinate axis Rz, the minimum sum of the coordinate values ​​in the direction of the first coordinate axis Rx and the coordinate values ​​in the direction of the third coordinate axis Rz, or the minimum other parameters generated after the two coordinate values ​​are subjected to different weightings or other operations. Based on the establishment of the above-mentioned model coordinate system, after comprehensively considering the coordinate values ​​in the directions of the first coordinate axis Rx and the third coordinate axis Rz, the vertex of the mesial incisor on the crown of the target tooth position is usually selected as the pole, especially the above-mentioned vertex of the central incisor is selected as the pole. When only considering the coordinate values ​​in the direction of the third coordinate axis Rz, the fifth spatial coordinate point m5 may also be selected as the pole. In particular, when considering the special case that the first tooth position 11a is blocked, it may also be selected. Fig.10 In order to illustrate the universal applicability of the present invention, the pole selection will be described below with consideration of the special case of occlusion, but the selection of other spatial points will not take into account the special case of occlusion.

[0158] A polar coordinate system is established with the pole m0, the horizontal axis of the polar coordinate system can be parallel to the first coordinate axis Rx to form a first polar coordinate horizontal axis Px, and the vertical axis can be parallel to the third coordinate axis Rz. The horizontal axis and the vertical axis of the polar coordinate system can be set to the same positive direction as the first coordinate axis Rx and the third coordinate axis Rz, that is, the first polar coordinate horizontal axis Px extends from the mesial surface to the distal surface of the crown of the target tooth position, and the vertical axis of the polar coordinate system extends from the incisal end of the crown of the target tooth position to the gingiva.

[0159] The polar angle and polar diameter are arranged in ascending order, and it can be preferred that the polar angle is preferentially arranged in ascending order, and when two spatial points have the same polar angle, the two are arranged in ascending order according to the polar diameter. For example, the sixth spatial point m6 and the seventh spatial point m7 have the same polar angle, but because the polar diameter of the sixth spatial point m6 is shorter than that of the seventh spatial point m7, the sixth spatial point m6 is arranged before the seventh spatial point m7. For another example, the polar diameter of the fourth spatial point m4 is shorter than that of the third spatial point m3, but because the polar angle of the fourth spatial point m4 is greater than that of the third spatial point m3, the third spatial point m3 is arranged before the fourth spatial point m4. Thus, at least a feature traversal sequence including the first spatial point m1, the second spatial point m2, the third spatial point m3, the fourth spatial point m4, the fifth spatial point m5, the sixth spatial point m6 and the seventh spatial point m7 can be formed in sequence, which can be expressed as .

[0160] Step 53, extract the coordinates of the first two spatial points in the feature traversal sequence, take the pole as the starting point, calculate the first polar coordinate vector and the second polar coordinate vector in sequence, and calculate the cross product of the first polar coordinate vector and the second polar coordinate vector to determine whether the cross product is less than 0.

[0161] The first polar coordinate vector can be expressed as , the second polar coordinate vector can be expressed as , the cross product can be expressed as If the cross product , it can be considered that the direction of the line segment m0m2 obtained by rotating the line segment m0m1 is counterclockwise. At this time, it can be judged that the first spatial point m1 and the second spatial point m2 may exist at the edge of the crown of the target tooth at the same time, that is, they belong to a convex hull reference point in the convex hull contour; but if the cross product , it can be considered that the direction of the line segment m0m2 obtained by rotating the line segment m0m1 is clockwise. At this time, it is necessary to further check whether the first spatial point m1 and the second spatial point m2 contain a convex hull reference point.

[0162] If yes, jump to step 54A and update the starting point to the first spatial point corresponding to the first polar coordinate vector;

[0163] If not, jump to step 54B to update the starting point to the second spatial point corresponding to the second polar coordinate vector.

[0164] When the counterclockwise rotation direction is met, it can be determined that the first spatial point m1 is closer to the pole m0 relative to the second spatial point m2, and the arrangement of the first spatial point m1 and the second spatial point m2 meets the arrangement of the convex hull reference point on the convex hull contour data relative to the pole m0, thereby determining the first spatial point m1 as the next starting point, and continuing to determine whether the second spatial point m2 and subsequent spatial points are within the convex hull contour data range. When the counterclockwise rotation direction is not met, it can be determined that there may be errors in the point selection and other steps, and the first spatial point m1 does not belong to the convex hull reference point in the convex hull contour data, so the traversal judgment is continued with the second spatial point m2 as the next starting point.

[0165] Step 55, traverse all other spatial points after the second spatial point in the feature traversal sequence, selectively update the starting point based on the polar coordinate vector formed by the spatial point and the starting point, and the cross product between the polar coordinate vectors, and determine at least two spatial points that meet the judgment conditions as convex hull reference points.

[0166] After selecting the next starting point after the pole m0, the second spatial point m2 and all other spatial points after it can be used as objects to perform cross product judgment, or all other spatial points after the second spatial point m2 can be used as objects to perform cross product judgment, and finally at least two convex hull reference points that meet the conditions and can be considered to fall into the convex hull contour data are obtained.

[0167] Step 56, fitting the convex hull contour data of the target tooth position according to the convex hull reference points and poles, and calculating the reference feature coordinates of the target tooth position according to the convex hull contour data.

[0168] At least two convex hull reference points and the pole m0 can form convex hull contour data that can be enclosed or fitted to form a convex hull. The larger the data amount of the convex hull reference points, the more the enclosed convex hull fits the contour of the target tooth crown. Based on this, preferably, when the coordinates of the midpoint of the gingival margin of the target tooth crown are selected as the reference feature coordinates, a convex hull reference point or a convex hull contour fitting point with the largest coordinate value on the first coordinate axis Rx can be selected as the reference feature coordinates. The convex hull contour fitting point can be obtained by executing the following steps: interpolating and fitting the convex hull contour data composed of the poles and the convex hull reference points to form a convex hull contour curve; wherein the convex hull contour curve includes convex hull reference points and convex hull contour fitting points obtained by interpolation. For example, when the first spatial point m1 and the second spatial point m2 are both convex hull reference points, a first convex hull contour fitting point can be interpolated between the pole m0 and the first spatial point m1, and / or a second convex hull contour fitting point can be interpolated between the first spatial point m1 and the second spatial point m2, thereby obtaining a smoother convex hull contour curve.

[0169] like Figure 7 , Figure 8 and Fig. 9 As shown, as another specific example independent of the above specific example, or as a preferred implementation of step 55 in the above specific example, step 55 may also specifically include the following steps. It can be understood that: Figure 7 Steps 410 to 43 can be combined with the above description to form a technical solution; Figure 8 Steps 51 to 56 are Figure 7 A part of step 411 in the above description may be combined to form another technical solution; Fig. 9 Steps 551 to 553 are as follows: Figure 8 as part of step 55 and further as Figure 7 Part of step 411 in the above description can be combined to form another technical solution. It is worth noting that this specific example includes steps 410 to 43, and steps 51 to 56, but they will not be described in detail below.

[0170] Step 551, extract the third spatial point located after the second spatial point in the feature traversal sequence, calculate the reference polar coordinate vector between the reference spatial point that is not the starting point in the first spatial point and the second spatial point and the starting point, and the third polar coordinate vector between the third spatial point and the starting point, calculate the cross product of the reference polar coordinate vector and the third polar coordinate vector, and determine whether the cross product is less than 0.

[0171] Combination Figure 6 and Fig.10 As shown, for example, in determining the first polar coordinate vector With the second polar coordinate vector The cross product , the first spatial point m1 is selected as the starting point after the pole m0, and the second spatial point m2 is defined as the reference spatial point. At this time, further calculation is performed as the reference polar coordinate vector With the third polar coordinate vector The cross product of , and determine the cross product Whether it is less than 0 is used to determine the rotation direction from line segment m1m2 to line segment m1m3.

[0172] If yes, jump to step 552A, update the starting point to the reference space point, and determine the reference space point to be the convex hull reference point.

[0173] If not, jump to step 552B, do not update the starting point, delete the reference space point, use the third space point as the new reference space point, and selectively update the starting point and determine the convex hull reference point based on the new reference polar coordinate vector and the cross product of the polar coordinate vector formed by the starting point and the next space point of the new reference space point.

[0174] If the cross product , then determine the second spatial point m2 as the reference spatial point as the new starting point after the first spatial point m1, and determine the second spatial point m2 as one of the convex hull reference points. , then delete the second spatial point m2 as the reference spatial point, take the first spatial point m1 as the starting point, and calculate the third polar coordinate vector The polar coordinate vector formed by the first spatial point m1 and the fourth spatial point m4 The cross product CP between them is iterated and judged according to the subsequent cross product results to determine whether to update the starting point and the convex hull reference point.

[0175] Step 553, repeat the iteration until all spatial points in the feature traversal sequence are judged and at least two convex hull reference points are obtained.

[0176] The following will describe the implementation process of the fourth spatial point m4, the fifth spatial point m5, the sixth spatial point m6 and the seventh spatial point m7. Figure 6 and Fig.10 Taking the case shown in as an example, the second spatial point m2 can be determined as the starting point after the first spatial point m1. Based on this, the third spatial point m3 and the fourth spatial point m4 are extracted to form the fourth polar coordinate vectors respectively. and the fifth polar coordinate vector , and use the third spatial point m3 as the reference spatial point to calculate the corresponding cross product And determine its relationship with 0, because the cross product Therefore, the third space point m3 is updated as the new starting point, and the third space point m3 is determined as the convex hull reference point.

[0177] Further extract the fourth spatial point m4 and the fifth spatial point m5, and form the sixth polar coordinate vector and the seventh polar coordinate vector , and use the fourth spatial point m4 as the reference spatial point to calculate the corresponding cross product And determine its relationship with 0, because the cross product Therefore, the starting point is not updated, the fourth spatial point m4 as the reference spatial point is deleted, the fifth spatial point m5 is used as the new reference spatial point, and the sixth spatial point m6 next to the fifth spatial point m5 is extracted to form the eighth polar coordinate vector , based on the seventh polar coordinate vector as the new reference polar coordinate vector , and the eighth polar coordinate vector The cross product of The relationship with 0 determines whether to update the starting point or determine the convex hull reference point.

[0178] because Therefore, the fifth spatial point m5 is updated as the new starting point, and the fifth spatial point m5 is determined as the convex hull reference point. The sixth spatial point m6 and the seventh spatial point m7 are further extracted to form the ninth polar coordinate vector and the tenth polar coordinate vector , and use the sixth spatial point m6 as the reference spatial point to calculate the corresponding cross product And determine its relationship with 0. Since the cross product Therefore, the starting point is not updated, the sixth spatial point m6 as the reference spatial point is deleted, and the seventh spatial point m7 is used as the new reference spatial point.

[0179] After the above steps, at least the first spatial point m1, the second spatial point m2, the third spatial point m3 and the fifth spatial point m5 are extracted as the convex hull reference points. The above technical solution is iterated repeatedly until all spatial points in the feature traversal sequence are judged, and all convex hull reference points for the target tooth position are obtained, thereby forming the corresponding convex hull contour and its data.

[0180] Although the above process forms a loop nesting algorithm in terms of expression, it can be implemented through algorithms such as stack and stack top pop in actual operation, and the present invention does not enumerate the specific operation methods.

[0181] like Figure 7 and Fig.11 As shown, the present invention provides a second embodiment of the method for generating oral structure information based on the above-mentioned embodiment, and the second embodiment refines step 412 and specifically provides steps 4121A and 4122A. It can be understood that other steps except the refinement step will not be described in detail below. The second embodiment specifically includes the following steps.

[0182] Step 410, obtaining oral cavity three-dimensional model data.

[0183] Step 411, determining at least one tooth position on the oral three-dimensional model as a target tooth position, and calculating the reference feature coordinates of the target tooth position based on a preset feature recognition rule to characterize the target tooth position features.

[0184] Step 4121A, draw a reference extension line along the first direction away from the target tooth position with the reference feature coordinates as the starting point, and continuously analyze the formation of the projection endpoints of the reference extension end on the maxillary surface of the oral three-dimensional model except the reference feature coordinates on the reference extension line.

[0185] The reference extension line can be an auxiliary line actually displayed on the oral three-dimensional model, so as to present the process of retrieving and determining the boundary feature coordinates to the medical staff for monitoring. Of course, the reference extension line can also be interpreted as the internal logic of the traversal process when the system executes the oral structure information generation method, that is, traversing the spatial points on the oral three-dimensional model in a direction similar to the reference extension line and judging the formation of the endpoints. In summary, the reference extension line is not required to actually exist, but exists in the oral structure information generation method provided by the present invention as an internal logic such as an auxiliary line or a direction reference line.

[0186] Combination Fig.12 and Fig.13 As shown, taking the fifth tooth position 11e as the target tooth position as an example, if the coordinates of the fifth gingival margin midpoint 11E corresponding to the fifth tooth position 11e are used as the reference feature coordinates, when the first direction is the extension direction of the tooth long axis, a fifth standard reference extension line Le will be generated accordingly, wherein one end of the fifth standard reference extension line Le is the fifth gingival margin midpoint 11E, and the projection of the other end on the oral three-dimensional model 100 corresponds to the projection endpoint. Of course, when the first direction is defined as the extension direction of the tooth long axis of the target tooth position or the projection of the extension direction of the tooth long axis on the jaw surface of the oral three-dimensional model 100, the corresponding reference extension line will form a certain inclination angle with the fifth standard reference extension line Le. The above three implementations are all feasible and can form the required projection endpoints. For the last technical solution, the first direction corresponds to the fifth projection direction D1e, and the fifth reference extension line Le' can be constructed; when the target tooth position is in the mandible, it can correspond to the fifth projection reverse direction D1e' opposite to the fifth projection direction D1e.

[0187] Similarly, taking the sixth tooth position 11f as the target tooth position as an example, if the sixth tooth position 11f is taken as the target tooth position, the corresponding sixth gingival margin midpoint 11F is included. When the first direction is the extension direction of the long axis of the tooth, the sixth standard reference extension line Lf will be generated through the above technical solution. When the first direction is the projection of the extension direction of the long axis of the tooth on the maxillary surface of the oral three-dimensional model 100, the sixth reference extension line Lf' will be generated through the above technical solution, and the corresponding first direction is the sixth projection direction D1f. When the target tooth position is in the mandible, the first direction corresponds to the sixth projection reverse direction D1f'.

[0188] It can be understood that although the present invention will determine the reference feature coordinates as the gingival margin midpoint coordinates of the target tooth position below, based on the foregoing description, those skilled in the art can derive a technical solution when determining the reference feature coordinates as other coordinates on the target tooth position.

[0189] Step 4122A, when the reference extension end no longer forms a projection endpoint with the maxillofacial surface of the oral three-dimensional model, the last formed projection endpoint coordinates are determined as the boundary feature coordinates.

[0190] In the process of any of the above-mentioned reference extension lines gradually extending in the direction away from the target tooth position, the reference extension line will at least continue to generate projection endpoints on the oral three-dimensional model 100. When the reference extension line extends beyond the boundary of the oral three-dimensional model 100, it cannot continue to generate the projection endpoints with the oral three-dimensional model 100. Based on this, according to the formation of the projection endpoints, when no new projection endpoints are formed, it is considered that the last formed projection endpoint is already on the boundary of the oral three-dimensional model 100.

[0191] For example, the fifth reference extension line Le' or the fifth standard reference extension line Le gradually extends from the fifth gingival margin midpoint 11E as the starting point, and continuously generates projection endpoints on the oral three-dimensional model 100, and the last projection endpoint is the corresponding fifth model boundary point 10E. For another example, the sixth reference extension line Lf' or the sixth standard reference extension line Lf gradually extends from the sixth gingival margin midpoint 11F as the starting point, and continuously generates projection endpoints on the oral three-dimensional model 100, and the last projection endpoint is the corresponding sixth model boundary point 10F.

[0192] Step 413, calculate and obtain the characteristic distance value between the reference characteristic coordinates and the boundary characteristic coordinates to characterize the positional relationship between the target tooth position and the model boundary.

[0193] Step 421 , compare the characteristic distance value with the distance integrity criterion value representing the preset position condition.

[0194] Step 422, judging whether the intraoral tissue region corresponding to the target tooth position in the oral three-dimensional model satisfies a preset integrity condition based on the comparison result.

[0195] If not satisfied, jump to step 43, obtain the corresponding intraoral image data, extract the oral feature data corresponding to the intraoral tissue area in the intraoral image data, establish the pixel size relationship between the intraoral image data and the oral three-dimensional model data, and reconstruct the part representing the oral feature data on the oral three-dimensional model data to obtain the oral structure information.

[0196] In this way, based on the above technical solution, not only can the boundary position of the model be determined based on a simple algorithm, avoiding the problems of excessive data volume, long calculation time and difficult fitting process caused by analyzing point cloud data, but it can also be based on tooth position features. Actively establish model boundary features corresponding to the tooth position features, which facilitates subsequent judgment of integrity and determination of missing positions.

[0197] In one application scenario, preferably, the intraoral tissue area includes the vestibule groove, the oral feature data includes the vestibule groove height data, the first direction is the projection of the extension direction of the tooth long axis on the jaw surface of the oral three-dimensional model, and the target tooth position includes the maxillary incisor tooth position and the mandibular incisor tooth position. In this way, based on the anatomical characteristics of the vestibule groove, the highest point of the upper vestibule groove is usually located above the maxillary central incisor or lateral incisor along the direction of the tooth long axis, and the lowest point of the lower vestibule groove is usually located below the mandibular central incisor or lateral incisor along the direction of the tooth long axis, and the corresponding model boundary features can be selected to determine whether the oral three-dimensional model includes the area necessary for calculating the vestibule groove height data. It is worth emphasizing that the incisor tooth position includes the central incisor tooth position and the lateral incisor tooth position, and further includes four directions: maxillary left side, mandibular left side, maxillary right side, and mandibular right side.

[0198] In another application scenario, preferably, the intraoral tissue area includes a vestibule groove, or a tooth root ridge, or includes both a vestibule groove and a tooth root ridge. The oral feature data includes dental arch width data, the first direction is the projection of the extension direction of the tooth long axis on the jaw surface of the oral three-dimensional model, and the target tooth position includes the first jaw surface left molar tooth position and the first jaw surface right molar tooth position corresponding to each other. Among them, the first jaw surface is the maxilla, or the mandible, or the maxilla and the mandible are selected as the first jaw surface at the same time. In this way, based on the anatomical features of the entire dental arch, the farthest point on the left side of the dental arch is usually located on the left side of the second molar on the left side of the maxilla along the projection direction of the tooth long axis, or on the left side of the second molar on the left side of the mandible along the projection direction of the tooth long axis, and the farthest point on the right side of the dental arch is usually located on the right side of the second molar on the right side of the maxilla along the projection direction of the tooth long axis, or on the right side of the second molar on the right side of the mandible along the projection direction of the tooth long axis, so as to select the corresponding model boundary features to determine whether the oral three-dimensional model includes the area necessary for calculating the dental arch width data.

[0199] In another application scenario, preferably, the oral tissue area includes the frenulum of the lip, and the oral characteristic data includes the frenulum of the lip width data. Figure 7 and Fig.14As shown, the present invention provides a third embodiment of the method for generating oral structure information based on the above-mentioned embodiment. The third embodiment refines step 411 and step 412 to adapt to the above-mentioned application scenario, and specifically provides step 411B, step 4121B and step 4122B. It can be understood that other steps except the refinement step will not be described in detail below. The third embodiment specifically includes the following steps.

[0200] Step 410, obtaining oral cavity three-dimensional model data.

[0201] Step 411B, determine the left central incisor tooth position of the first maxillary face and the right central incisor tooth position of the first maxillary face on the three-dimensional oral model as the target tooth position, and based on the preset feature recognition rules, calculate the reference feature coordinates of the target tooth position and several relative feature coordinates on the target tooth position located between the two reference feature coordinates to characterize the target tooth position characteristics.

[0202] Wherein, the first maxillary surface is the maxilla, or the mandible, or both the maxilla and the mandible are selected as the first maxillary surface. Fig.14 As shown, taking the first maxillary face as the maxilla as an example, the right central incisor tooth position of the first maxillary face points to the fifth tooth position 11e, and the left central incisor tooth position of the first maxillary face points to the seventh tooth position 11g, then the coordinates of the fifth gingival margin midpoint 11E can be used as the reference feature coordinates of the tooth position, and the coordinates of the seventh gingival margin midpoint 11G can be used as the reference feature coordinates of the tooth position. Since the central incisors located on the same maxillary face are arranged in sequence along a direction perpendicular to the tooth midline or the long axis of the tooth body, similar to extracting crown edge points such as convex hull reference points, the fifth gingival margin midpoint 11E and the seventh gingival margin midpoint 11G must include multiple spatial points distributed on the target tooth position. As a preferred embodiment, the coordinates of a spatial point that is simultaneously located between the fifth gingival margin midpoint 11E and the seventh gingival margin midpoint 11G and distributed on the gingival margin of the fifth tooth position 11e or the gingival margin of the seventh tooth position 11g can be selected as the relative feature coordinates. It can be understood that the spatial point pointed to by the relative characteristic coordinates is also located on the target tooth position, so the relative characteristic coordinates and the reference characteristic coordinates can jointly characterize the characteristics of the target tooth position.

[0203] For example, the coordinates of the fifth gingival margin marking point 11Em located on the fifth tooth position 11e can be selected as the relative characteristic coordinates. Of course, the spatial point pointed to by the relative characteristic coordinates is not limited to the fifth gingival margin marking point 11Em, and can be any point located on the gingival margin or any point located on the tooth crown.

[0204] Step 4121B, respectively, draw a reference extension line along the first direction away from the target tooth position with the reference feature coordinates and the relative feature coordinates as starting points, and continuously analyze the formation of the projection endpoints of the reference extension ends on the reference extension line other than the reference feature coordinates or the relative feature coordinates on the oral three-dimensional model.

[0205] Among them, the first direction is the projection of the extension direction of the long axis of the tooth on the jaw surface of the three-dimensional model of the oral cavity. Based on this, corresponding to the fifth tooth position 11e and the fifth gingival margin midpoint 11E, a fifth reference extension line Le' can be generated, corresponding to the seventh tooth position 11g and the seventh gingival margin midpoint 11G, a seventh reference extension line Lg' can be generated, and corresponding to the fifth tooth position 11e and the fifth gingival margin marking point 11Em, a fifth marked reference extension line Lem' can be generated as one of the reference extension lines. Similarly, the target tooth position corresponding to each of the above-mentioned reference extension lines is in the maxilla, and the first direction D1 is from the incisal ridge of the central incisor tooth position upward; when the target tooth position is selected as the central incisor in the mandible, the extension direction on which the corresponding reference extension line relies can be the first reverse direction D1' (for example, any of the above-mentioned reference reverse directions).

[0206] Step 4122B, when the reference extension end no longer forms a projection endpoint with the oral three-dimensional model, the last formed projection endpoint coordinates are determined as the boundary feature coordinates.

[0207] The projection endpoints are continuously generated on the oral three-dimensional model 100 during the extension of the reference extension lines. When the at least three reference extension lines no longer generate new projection endpoints with the oral three-dimensional model 100, it is considered that the boundary of the oral three-dimensional model 100 has been touched. Based on this, the fifth model boundary point 10E corresponding to the fifth tooth position 11e, the seventh model boundary point 10G corresponding to the seventh tooth position 11g, and the fifth marked boundary point 10Em corresponding to the fifth gingival margin marked point 11Em on the fifth tooth position 11e can be determined, so that the coordinates of the above model boundary points and the marked boundary points are used together as the boundary feature coordinates.

[0208] Step 413, calculate and obtain the characteristic distance value between the reference characteristic coordinates and the boundary characteristic coordinates to characterize the positional relationship between the target tooth position and the model boundary.

[0209] Step 421 , compare the characteristic distance value with the distance integrity criterion value representing the preset position condition.

[0210] Step 422, judging whether the intraoral tissue region corresponding to the target tooth position in the oral three-dimensional model satisfies a preset integrity condition based on the comparison result.

[0211] If not satisfied, jump to step 43, obtain the corresponding intraoral image data, extract the oral feature data corresponding to the intraoral tissue area in the intraoral image data, establish the pixel size relationship between the intraoral image data and the oral three-dimensional model data, and reconstruct the part representing the oral feature data on the oral three-dimensional model data to obtain the oral structure information.

[0212] The principle of the above technical solution is that, based on the anatomical characteristics of the lip frenulum, it is usually located in the middle area above the two central incisors in the maxilla, and in the middle area below the two central incisors in the mandible. Therefore, the central incisor tooth position can be selected as the target tooth position, and the range of the area where the lip frenulum is located can be limited to improve the calculation speed. Since the model boundary features corresponding to the relative feature coordinates are retrieved, it is possible to further determine whether the area of ​​the lip frenulum in the oral three-dimensional model is missing based on the positional relationship between the relative feature coordinates and the model boundary features. The technical solution for determining whether it is missing can use the same set of oral structure information generation methods as the aforementioned method for determining whether the vestibule groove is missing, which will not be repeated here.

[0213] It can be seen that on this basis, those skilled in the art can obtain the positional relationship between the target tooth position and the model boundary corresponding to different areas on the oral 3D model based on the anatomical characteristics of different oral tissues. The present invention does not exhaustively list the technical solutions corresponding to all oral tissues.

[0214] like Figure 7 and Fig.16 As shown, the present invention provides a fourth embodiment of the method for generating oral structure information based on the above-mentioned embodiment. The fourth embodiment provides pre-steps 61 to 62 for step 421A. Steps 61 to 62 can be set at any position before step 421A, and the present invention is not limited to this. Hereinafter, the present invention will be described by taking the fourth embodiment in which the pre-step is set before step 410 as an example. It can be understood that other steps except the pre-step will not be described in detail below. The fourth embodiment specifically includes the following steps.

[0215] Step 61: Obtain at least two sets of three-dimensional training model data and at least two sets of distance training data corresponding to the three-dimensional training model data.

[0216] The "preset location condition" can be directly obtained and set based on experiments and network data. Of course, the technical solution provided in the fourth embodiment of the present invention can also be used to obtain a more accurate and practical preset location condition.

[0217] In this embodiment, it is preferred to use three-dimensional training model data that at least includes corresponding features of the complete intraoral tissue area to calculate the preset position conditions. For example, when the generation of oral structure information focuses more on the vestibule groove, the three-dimensional training model data used to construct the preset position conditions preferably include all vestibule groove features and tooth position features, but for things such as the position and shape of the root ridge, missing parts are allowed. It should be noted at this point that although the data used to formulate the preset position conditions in the fourth embodiment is three-dimensional training model data, when only focusing on local features such as lip frenulum, maxillofacial protrusion amplitude, etc., it can also be partial or local intraoral image data in the three-dimensional training model. When applying planar image data, it is necessary to add steps to establish pixel size relationships and convert two-dimensional features into three-dimensional features.

[0218] The distance training data includes the distance between the conditional training coordinates corresponding to the reference feature coordinates on the three-dimensional training model and the target tissue coordinates corresponding to the intraoral tissue area on the three-dimensional training model. The line connecting the conditional training coordinates and the target tissue coordinates extends along the first direction.

[0219] like Figure 6 and Fig.17 As shown, taking the intraoral tissue area as the vestibule groove area and the first tooth position 11a corresponding to the maxillary left central incisor tooth position as an example, the reference feature coordinates corresponding to the first tooth position 11a in the actual oral three-dimensional model 100 point to the coordinates of the first gingival margin midpoint 11A, the conditional training coordinates corresponding to the maxillary left central incisor tooth position in the three-dimensional training model point to the coordinates of the first training gingival margin midpoint 1121U, and the target tissue coordinates corresponding to the maxillary left central incisor tooth position in the three-dimensional training model point to the coordinates of the first training vestibule groove bottom point 1321U. Based on this, the distance training data can be jointly determined according to the coordinates of the first training gingival margin midpoint 1121U and the coordinates of the first training vestibule groove bottom point 1321U.

[0220] The corresponding relationship between the above data is established based on the first gingival margin midpoint 11A and the first training gingival margin midpoint 1121U as references, respectively, and is established according to the first direction D1. Since there is a natural corresponding relationship between the gingival margin midpoints, the corresponding relationship can be quickly and stably established by traversing the first gingival margin midpoint 11A and the first training gingival margin midpoint 1121U retrieved according to the first direction D1.

[0221] The distance training data is a plurality of data formed for a single tooth position in a plurality of three-dimensional training model data. In other words, a distance training data corresponding to a first tooth position can be obtained in a first three-dimensional training model, and another distance training data corresponding to a second tooth position can be obtained in a second three-dimensional training model.

[0222] Step 62, calculate the average distance data and the training distance standard deviation of the distance training data, and calculate the distance integrity criterion value according to the difference between the product of the average distance data and the training distance standard deviation and the preset distribution probability coefficient.

[0223] The distribution probability coefficient points to a preset probability distribution, which can be a discrete distribution such as geometric distribution, binomial distribution, Poisson distribution, etc., or a continuous distribution such as uniform distribution, exponential distribution, normal distribution, etc. The distance integrity criterion value thus established is used to characterize the preset position conditions, and can estimate and cover various situations that may occur on the oral three-dimensional model to a greater extent, thereby reducing the probability of misjudgment.

[0224] Step 410, obtaining oral cavity three-dimensional model data.

[0225] Step 411, determining at least one tooth position on the oral three-dimensional model as a target tooth position, and calculating the reference feature coordinates of the target tooth position based on a preset feature recognition rule to characterize the target tooth position features.

[0226] Step 412: According to the reference feature coordinates, boundary feature coordinates corresponding to the reference feature coordinates on the oral cavity three-dimensional model are determined along the first direction to characterize the model boundary features.

[0227] Step 413, calculate and obtain the characteristic distance value between the reference characteristic coordinates and the boundary characteristic coordinates to characterize the positional relationship between the target tooth position and the model boundary.

[0228] Step 421 , compare the characteristic distance value with the distance integrity criterion value representing the preset position condition.

[0229] Step 422, judging whether the intraoral tissue region corresponding to the target tooth position in the oral three-dimensional model satisfies a preset integrity condition based on the comparison result.

[0230] If not satisfied, jump to step 43, obtain the corresponding intraoral image data, extract the oral feature data corresponding to the intraoral tissue area in the intraoral image data, establish the pixel size relationship between the intraoral image data and the oral three-dimensional model data, and reconstruct the part representing the oral feature data on the oral three-dimensional model data to obtain the oral structure information.

[0231] In this way, based on the three-dimensional training model data with a high degree of data integrity, a completeness criterion value with a high reference value is formed, which can better judge whether the regional part corresponding to the target tooth position has feature loss. At the same time, in the process of implementing the fourth embodiment, by adjusting the first direction, other positional relationships and preset position conditions can be established based on the actual oral three-dimensional model and the three-dimensional training model, so as to complete the regional integrity judgment of the intraoral tissue in dimensions such as dental arch width, dental arch curvature, maxillofacial protrusion amplitude and lip frenulum width.

[0232] In one application scenario, preferably, the intraoral tissue area includes the vestibule groove, the oral feature data includes the vestibule groove height data, the target tooth position includes the maxillary first incisor tooth position and the mandibular second incisor tooth position, and the distance training data includes a first groove bottom distance parameter corresponding to the maxillary first incisor tooth position, and a second groove bottom distance parameter corresponding to the mandibular second incisor tooth position. Based on this, on the basis of the above fourth embodiment, as Fig.16 As shown, the step 62 may specifically include the following steps as part thereof.

[0233] Step 621A, calculate the first trench bottom average distance value of all first trench bottom distance parameters in all distance training data, and the second trench bottom average distance value of all second trench bottom distance parameters in all distance training data, to obtain average distance data.

[0234] Step 622A, calculate the first groove bottom distance standard deviation of all first groove bottom distance parameters in all distance training data, and the second groove bottom distance standard deviation of all second groove bottom distance parameters in all distance training data, to obtain the training distance standard deviation.

[0235] by Fig.17 The three-dimensional training model data containing a set of distance training data is represented as an object, and the maxillary left central incisor is taken as the maxillary first incisor, and the mandibular right lateral incisor is taken as the mandibular second incisor as an example. The first groove bottom distance parameter points to the distance between the first training gingival margin midpoint 1121U and the first training vestibule groove bottom point 1321U distributed in sequence along the first direction D1 of the maxillary first incisor tooth position; the second groove bottom distance parameter points to the distance between the second training gingival margin midpoint 1142L and the second training vestibule groove bottom point 1342L distributed in sequence along the first direction D1 of the mandibular second incisor tooth position.

[0236] Based on this, the distance training data of all three-dimensional training model data (for example, there are n groups in total) can be integrated to obtain n groups of first groove bottom distance parameters corresponding to the maxillary left central incisor, and n groups of second groove bottom distance parameters corresponding to the mandibular right lateral incisor. From this, the average values ​​can be calculated as the first groove bottom average distance value and the second groove bottom average distance value, respectively, to form average distance data, and the standard deviations can be calculated as the first groove bottom distance standard deviation and the second groove bottom distance standard deviation, respectively, to form the training distance standard deviation.

[0237] In a preferred embodiment, the distance training data includes groove bottom distance parameters corresponding to the tooth positions of the maxillary left lateral incisor (ie, Fig.17 The distance between the fifth training gingival margin midpoint 1122U and the fifth training vestibule groove bottom point 1322U along the first direction D1), the groove bottom distance parameter of the maxillary left central incisor tooth position ( Fig.17 The distance between the midpoint of the first training gingival margin 1121U and the bottom point of the first training vestibule groove 1321U along the first direction D1), the groove bottom distance parameter of the maxillary right central incisor tooth position ( Fig.17 The distance between the fourth training gingival margin midpoint 1111U and the fourth training vestibule groove bottom point 1311U along the first direction D1), the groove bottom distance parameter of the maxillary right lateral incisor tooth position ( Fig.17 The distance between the third training gingival margin midpoint 1112U and the third training vestibule groove bottom point 1312U along the first direction D1), the groove bottom distance parameter of the mandibular left lateral incisor tooth position ( Fig.17 The distance between the eighth training gingival margin midpoint 1132L and the eighth training vestibule groove bottom point 1332L along the first direction D1), the groove bottom distance parameter of the mandibular left central incisor tooth position ( Fig.17 The distance between the midpoint of the seventh training gingival margin 1131L and the bottom point of the seventh training vestibule groove 1331L along the first direction D1), the groove bottom distance of the mandibular right central incisor tooth position ( Fig.17 The distance between the sixth training gingival margin midpoint 1141L and the sixth training vestibule groove bottom point 1341L along the first direction D1) and the groove bottom distance parameter of the mandibular right lateral incisor tooth position ( Fig.17 The distance between the second training gingival margin midpoint 1142L and the second training vestibule groove bottom point 1342L along the first direction D1). The first directions D1 corresponding to different target tooth positions may be different, and for the sake of simplicity of description, they are not specifically distinguished and marked. In this way, a more accurate and reliable distance integrity criterion value can be calculated using the groove bottom distance parameters of the eight incisor tooth positions.

[0238] It should be emphasized that the groove bottom distance parameter represents the distance between the conditional training coordinates of the gingival margin midpoint corresponding to the incisor tooth position and the target tissue coordinates of the groove bottom feature point corresponding to the gingival margin midpoint. Therefore, in other scenarios, the groove bottom distance parameters or other parameters of other target tooth positions can be calculated based on the same logic.

[0239] In another application scenario, preferably, the intraoral tissue area includes the vestibule groove, or the root ridge, or both the vestibule groove and the root ridge, the oral feature data includes the dental arch width data, the target tooth position includes the distal molar positions on both sides of the maxilla and the distal molar positions on both sides of the mandible, and the distance training data includes the upper dental arch width parameters corresponding to the distal molar positions on both sides of the maxilla, and the lower dental arch width parameters corresponding to the distal molar positions on both sides of the mandible. Based on this, on the basis of the fourth embodiment above, as Fig.16 As shown, the step 62 may specifically include the following steps as part thereof.

[0240] Step 621B, calculate the upper average width value of all upper dental arch width parameters in all distance training data, and the lower average width value of all lower dental arch width parameters in all distance training data, to obtain average distance data.

[0241] Step 622B, calculate the upper width standard deviation of all upper dental arch width parameters in all distance training data, and the lower width standard deviation of all lower dental arch width parameters in all distance training data, to obtain the training distance standard deviation.

[0242] The steps 621B and 622B correspond to the steps 621A and 621A, and the dental arch width parameter is used as the calculation object to judge the integrity of the vestibule groove or the root ridge. The basic idea is consistent with the previous application scenario, and will not be described in detail here.

[0243] The distal molar positions on both sides of the maxilla are preferably the second molar positions on the left side of the maxilla and the second molar positions on the right side of the maxilla; the distal molar positions on both sides of the mandible are preferably the second molar positions on the left side of the mandible and the second molar positions on the right side of the mandible. Of course, when there are missing teeth, multiple teeth (third molars protruding), etc., the tooth position farthest from the tooth midline can be selected as the "distal molar position".

[0244] For any specific example or derived technical solution in the fourth embodiment, it is further preferred that the distribution probability coefficient is 2.

[0245] For example, in an application scenario where the oral feature data includes vestibular groove height data, the first maxillary incisor tooth position is taken as the object, and the calculated average distance value of the first groove bottom is , the calculated standard deviation of the first ditch bottom distance is , then the distribution probability coefficient In a preferred embodiment of the present invention, the first distance integrity criterion value corresponding to the maxillary first incisor tooth position is , can be configured to at least satisfy:

[0246] .

[0247] Thus, the first characteristic distance value corresponding to the first incisor tooth position can be compared , and the first distance integrity criterion value , so as to achieve the effect of “comparing the position relationship with the preset position condition corresponding to the position relationship”.

[0248] In a preferred embodiment, corresponding to step 413 described in any technical solution herein, the step of "calculating the characteristic distance value between the reference characteristic coordinates and the boundary characteristic coordinates" can be preferably configured to include: calculating the distance value between the reference characteristic coordinates and the boundary characteristic coordinates in the extension direction of the long axis of the tooth to obtain the characteristic distance value.

[0249] Corresponds to Figure 6 For example, the characteristic distance value corresponding to the first tooth position 11a may be a distance value between the coordinates of the first gingival margin midpoint 11A and the coordinates of the first model boundary point 10A in the first reference direction D11.

[0250] Based on this, in one embodiment, the characteristic distance value calculation step can be further simplified by establishing a special coordinate system. In this embodiment, the first direction D1 can be defined as the projection of the extension direction of the long axis of the tooth on the oral three-dimensional model 100. Correspondingly, the boundary characteristic coordinates and the reference characteristic coordinates are located in a preset model coordinate system, and the model coordinate system at least includes a first coordinate axis Rx and a third coordinate axis Rz, the third coordinate axis Rz extends along the extension direction of the tooth centerline, and the first coordinate axis Rx extends along the extension direction of the central incisor width. At this time, the "calculation of the distance value between the reference characteristic coordinates and the boundary characteristic coordinates in the extension direction of the tooth centerline" can be specifically configured to include: calculating the coordinate difference between the reference characteristic coordinates and the boundary characteristic coordinates on the third coordinate axis. For example, the coordinate difference between the coordinates of the first gingival margin midpoint 11A and the coordinates of the first model boundary point 10A on the third coordinate axis Rz is calculated.

[0251] like Figure 7 and Fig.18As shown, the present invention provides a fifth embodiment of the method for generating oral structure information based on the above-mentioned embodiment. The fifth embodiment refines step 421, specifically provides step 4211, step 4212A and step 4212B, and refines step 422, providing step 4220 for step 4212A and step 4212B. It can be understood that other steps except the refinement step will not be described in detail below. Similarly, since steps 4211 to 4220 belong to step 421 and step 422 respectively, the description of step 421 and step 422 themselves will be omitted below. The fifth embodiment specifically includes the following steps.

[0252] Step 410, obtaining oral cavity three-dimensional model data.

[0253] Step 411, determining at least one tooth position on the oral three-dimensional model as a target tooth position, and calculating the reference feature coordinates of the target tooth position based on a preset feature recognition rule to characterize the target tooth position features.

[0254] Step 412: According to the reference feature coordinates, boundary feature coordinates corresponding to the reference feature coordinates on the oral cavity three-dimensional model are determined along the first direction to characterize the model boundary features.

[0255] Step 413, calculate and obtain the characteristic distance value between the reference characteristic coordinates and the boundary characteristic coordinates to characterize the positional relationship between the target tooth position and the model boundary.

[0256] Step 4211, determine whether the feature distance value is less than the distance integrity criterion value.

[0257] If so, jump to step 4212A to determine that the position in the oral three-dimensional model pointed to by the feature distance value is the feature missing position.

[0258] If not, jump to step 4212B and continue to compare the numerical values ​​of the next feature distance value with the distance integrity criterion value.

[0259] Step 4220, based on the number of feature missing positions, determine whether the intraoral tissue area corresponding to the target tooth position in the oral three-dimensional model meets the integrity condition.

[0260] If not satisfied, jump to step 43, obtain the corresponding intraoral image data, extract the oral feature data corresponding to the intraoral tissue area in the intraoral image data, establish the pixel size relationship between the intraoral image data and the oral three-dimensional model data, and reconstruct the part representing the oral feature data on the oral three-dimensional model data to obtain the oral structure information.

[0261] Based on the fifth embodiment, the feature missing positions can be determined in advance, and based on the number of feature missing positions, it is evaluated whether the oral cavity three-dimensional model is within the range that does not meet the integrity condition.

[0262] like Figure 6 As shown, the first characteristic distance value of the first tooth position 11a (that is, the first incisor tooth position, or the maxillary left central incisor tooth position) is continued as described above. , and the first distance integrity criterion value The definition of , then the first tooth position 11a is determined to be a feature missing position. Continuing with the fourth feature distance value of the fourth tooth position 11d (ie, the sixth incisor tooth position, or the mandibular right central incisor tooth position) in the previous text , and the fourth distance integrity criterion value The definition of , it is determined that the fourth tooth position 11d is not a feature missing position, so the next target tooth position is determined.

[0263] Figure 6 The figure shows the distribution of intraoral tissues corresponding to eight incisor positions in the maxillary and mandibular teeth. It can be seen that the positions corresponding to the maxillary left lateral incisor, the maxillary left central incisor and the maxillary right lateral incisor are feature missing positions. At this time, the "number of feature missing positions" corresponds to 3. Based on this, a fixed threshold or a dynamic threshold can be set for the number to further determine whether the intraoral tissue area corresponding to the vestibule groove height pointed to by the above eight incisor positions meets the integrity condition.

[0264] The specific judgment rule of the integrity condition provided by step 4220 can be specifically configured to include the following steps in a specific example based on this embodiment. However, it can be understood that the judgment rule is not limited to the fifth embodiment, and those skilled in the art are capable of combining the judgment rule with step 42, step 422 or step 4220 in other implementation modes, embodiments or specific examples to form a preferred technical solution of the corresponding embodiment.

[0265] Step 4221, determine the numerical relationship between the number of feature missing positions and the allowable error value.

[0266] If the number of the feature missing positions is greater than or equal to the allowable error value, jump to step 4222A to determine that the oral tissue area corresponding to the target tooth position in the oral three-dimensional model does not meet the integrity condition.

[0267] If the number of the feature missing positions is less than the allowable error value, jump to step 4222B to determine whether the intraoral tissue area corresponding to the target tooth position in the oral three-dimensional model meets the integrity condition.

[0268] In this way, it is possible to directly determine whether the intraoral tissue area meets the integrity condition based on the comparison of the preset allowable error value with the number of feature missing positions, thereby converting the abstract concept of integrity condition into data content that can be actually calculated.

[0269] For any technical solution including the above specific examples, it is further preferred that the allowable error value is an integer greater than or equal to half of the number of target tooth positions. In this way, the allowable error value is set as a threshold value that changes dynamically with the number of target tooth positions, which can reflect the ratio of the number of feature missing positions to the number of target tooth positions whose feature distance values ​​are calculated, and control the requirements for the integrity of the oral three-dimensional model from a macro perspective.

[0270] For example, if eight tooth positions, including the maxillary left and right central incisor positions, the maxillary left and right lateral incisor positions, the mandibular left and right central incisor positions, and the mandibular left and right lateral incisor positions, are selected as the target tooth positions, then when the number of feature missing positions is greater than or equal to 4, it is determined that the target tooth positions correspond to the intraoral tissue area of ​​the oral three-dimensional model that does not meet the integrity condition. Figure 6 In the oral three-dimensional model shown, there are three corresponding feature missing positions at the maxillary left lateral incisor, the maxillary left central incisor and the maxillary right lateral incisor, which is less than half of the total number of target tooth positions. Therefore, it can be considered that the completeness condition is met.

[0271] Preferably, after determining that the intraoral tissue area corresponding to the target tooth position of the oral three-dimensional model data meets the integrity condition, the oral three-dimensional model itself can also be further subjected to feature extraction and fitting. For feature extraction, since the intraoral tissue areas corresponding to the maxillary right central incisor tooth position and the mandibular right central incisor tooth position and the mandibular right lateral incisor tooth position all contain relatively complete vestibule groove features, the vestibule groove height data contained in the oral feature data can be obtained accordingly, that is, the oral feature data is extracted according to the non-feature missing position. For fitting, the structures on the left and right sides of the maxillary face are usually symmetrical to each other, so at least the vestibule groove features corresponding to the maxillary right central incisor tooth position can be supplemented into the feature missing position corresponding to the maxillary left central incisor tooth position after mirroring and other processing, that is, the oral structure information is reconstructed based on the oral anatomical features according to the non-feature missing position.

[0272] like Fig.19As shown, another embodiment of the present invention provides a method for generating oral structure information. The application or instruction corresponding to the method can be installed in the above storage medium and / or the above oral structure information generation system 300 to achieve the technical effect of generating oral structure information. This further embodiment mainly refines step 41 and step 42 in the previous embodiment, and provides steps 410 to 413' corresponding to step 41, and provides steps 421' to 422 corresponding to step 42. The oral structure information generation method can specifically include the following steps.

[0273] Step 410, obtaining oral cavity three-dimensional model data.

[0274] Step 411 ′, determining at least the incisor tooth position on the oral three-dimensional model as the target tooth position, and calculating the reference feature coordinates of the target tooth position based on a preset feature recognition rule to characterize the target tooth position feature.

[0275] Step 412': determine the upper boundary extreme value coordinates and the lower boundary extreme value coordinates of the oral cavity three-dimensional model along the first direction according to the reference feature coordinates.

[0276] Step 413 ′, calculate and obtain the characteristic height value between the upper boundary extreme value coordinates and the lower boundary extreme value coordinates to characterize the positional relationship between the maxillary model boundary and the mandibular model boundary.

[0277] Step 421', compare the characteristic height value with the height integrity criterion value representing the preset position condition, and judge whether the intraoral tissue area corresponding to the target tooth position in the oral three-dimensional model meets the preset integrity condition based on the comparison result.

[0278] Step 422, judging whether the intraoral tissue region corresponding to the target tooth position in the oral three-dimensional model satisfies a preset integrity condition based on the comparison result.

[0279] If not satisfied, jump to step 43, obtain the corresponding intraoral image data, extract the oral feature data corresponding to the intraoral tissue area in the intraoral image data, establish the pixel size relationship between the intraoral image data and the oral three-dimensional model data, and reconstruct the part representing the oral feature data on the oral three-dimensional model data to obtain the oral structure information.

[0280] Among them, the upper boundary extreme value coordinates are located in the upper jaw of the oral three-dimensional model, and the distance between the upper boundary and the occlusal plane has a maximum value, and the lower boundary extreme value coordinates are located in the lower jaw of the oral three-dimensional model, and the distance between the upper boundary and the occlusal plane has a maximum value.

[0281] The above-mentioned further embodiment provided by the present invention provides a technical solution that is different from the above-mentioned embodiment and uses the positional relationship between the maxillary model boundary and the mandibular model boundary to determine whether the integrity condition is met. For the steps of determining the target tooth position and calculating the reference feature coordinates, the technical solution provided in the above text can be applied interchangeably. Furthermore, other further improved technical solutions described in the previous embodiment can also be applied interchangeably in this further embodiment after appropriate adjustments, thereby forming more derived technical solutions.

[0282] Combination Fig.21 As shown, in another embodiment, at least eight incisor tooth positions are determined as target tooth positions, and the maxillary model boundary features and the mandibular model boundary features are traversed and retrieved according to the corresponding at least eight reference feature coordinates, and finally the upper boundary extreme value coordinate point 101M and the lower boundary extreme value coordinate point 102M, which are farthest from the occlusal plane, are determined. Due to the positional relationship between the upper boundary extreme value coordinate point 101M and the lower boundary extreme value coordinate point 102M, the positional relationship between the maxillary model boundary and the mandibular model boundary can be used to characterize the positional relationship between the maxillary model boundary and the mandibular model boundary, so that the positional relationship between the abstract boundaries can be converted into the positional relationship between the specific coordinate points, which is convenient for algorithm design and judgment.

[0283] like Fig.19 and Fig. 20 As shown, the present invention provides a first embodiment of the method for generating oral structure information based on the above-mentioned embodiment. The first embodiment refines step 412', specifically provides step 4121' and step 4122', and refines step 413', providing step 4130' corresponding to step 4121' to step 4122'. It can be understood that other steps except the refinement step will not be described in detail below. Similarly, since steps 4121' to 4130' belong to step 412' and step 413' respectively, the description of step 412' and step 413' themselves will be omitted below. The first embodiment specifically includes the following steps.

[0284] Step 410, obtaining oral cavity three-dimensional model data.

[0285] Step 411 ′, determining at least the incisor tooth position on the oral three-dimensional model as the target tooth position, and calculating the reference feature coordinates of the target tooth position based on a preset feature recognition rule to characterize the target tooth position feature.

[0286] Step 4121', calculate the upper boundary coordinate set corresponding to the maxillary target tooth position on the oral 3D model, and the lower boundary coordinate set corresponding to the mandibular target tooth position on the oral 3D model, to characterize the model boundary features.

[0287] Step 4122', traverse and determine the upper boundary coordinate with the maximum coordinate value corresponding to the third coordinate axis in the upper boundary coordinate set as the upper boundary extreme value coordinate, traverse and determine the lower boundary coordinate with the minimum coordinate value corresponding to the third coordinate axis in the lower boundary coordinate set as the lower boundary extreme value coordinate.

[0288] Step 4130', calculating the coordinate difference between the upper boundary extreme value coordinate and the lower boundary extreme value coordinate on the third coordinate axis as the characteristic height value.

[0289] Step 421', compare the characteristic height value with the height integrity criterion value representing the preset position condition, and judge whether the intraoral tissue area corresponding to the target tooth position in the oral three-dimensional model meets the preset integrity condition based on the comparison result.

[0290] Step 422, judging whether the intraoral tissue region corresponding to the target tooth position in the oral three-dimensional model satisfies a preset integrity condition based on the comparison result.

[0291] If not satisfied, jump to step 43, obtain the corresponding intraoral image data, extract the oral feature data corresponding to the intraoral tissue area in the intraoral image data, establish the pixel size relationship between the intraoral image data and the oral three-dimensional model data, and reconstruct the part representing the oral feature data on the oral three-dimensional model data to obtain the oral structure information.

[0292] Among them, the first direction is the projection of the extension direction of the long axis of the tooth on the maxillary surface of the oral three-dimensional model, the reference feature coordinates, the upper boundary extreme value coordinates and the lower boundary extreme value coordinates are located in a preset model coordinate system, the origin of the model coordinate system is located on the straight line where the incisor ridge is located, and at least includes a first coordinate axis and a third coordinate axis, the third coordinate axis extends along the extension direction of the tooth midline, and the first coordinate axis extends along the extension direction of the central incisor width.

[0293] Combination Figure 6 and Fig.21 As shown, the first coordinate axis is Rx, the third coordinate axis is Rz, and the origin Ro of the model coordinate system is located on the straight line where the incisor ridge of the central incisor or the lateral incisor is located, and is preferably located between the mesial incisor angle of the maxillary left central incisor and the mesial incisor angle of the maxillary right central incisor. Of course, the origin Ro of the model coordinate system can also be interpreted as preferably located at the intersection of the tooth midline and the occlusal plane.

[0294] In this first embodiment, the boundary feature coordinates farthest from the occlusal plane are determined by traversing all the upper boundary coordinates and all the lower boundary coordinates. When the target tooth position is determined to be the incisor, or the target intraoral tissue area is determined to be the vestibule groove area, the eight boundary feature coordinates corresponding to the eight incisor tooth positions can be determined based on the above technical solution. Furthermore, the coordinates of the spatial point with the largest coordinate value on the third coordinate axis Rz among the four boundary feature coordinates located in the maxilla are selected as the upper boundary extreme value coordinates 101M, and the coordinates of the spatial point with the smallest coordinate value on the third coordinate axis Rz among the four boundary feature coordinates located in the mandible are selected as the lower boundary extreme value coordinates 102M, thereby calculating the coordinate difference between the two on the third coordinate axis Rz as the characteristic height value. .

[0295] like Fig.19 and Fig. 22 As shown, the present invention provides a second embodiment of the method for generating oral structure information based on the above-mentioned embodiment. The second embodiment provides pre-steps 61' to 62' for step 421'. Steps 61' to 62' can be set at any position before step 421', and the present invention is not limited to this. Hereinafter, the present invention will be described in detail by taking the second embodiment in which the pre-step is set before step 410 as an example. It can be understood that other steps except the pre-step will not be described in detail below. The second embodiment specifically includes the following steps.

[0296] Step 61 ′: acquiring at least two sets of three-dimensional training model data and at least two sets of height training data corresponding to the three-dimensional training model data.

[0297] Step 62', calculate the average height data and the training height standard deviation of the height training data, and calculate the height integrity criterion value according to the difference between the product of the average height data and the training height standard deviation and the preset distribution probability coefficient.

[0298] Step 410, obtaining oral cavity three-dimensional model data.

[0299] Step 411 ′, determining at least the incisor tooth position on the oral three-dimensional model as the target tooth position, and calculating the reference feature coordinates of the target tooth position based on a preset feature recognition rule to characterize the target tooth position feature.

[0300] Step 412': determine the upper boundary extreme value coordinates and the lower boundary extreme value coordinates of the oral cavity three-dimensional model along the first direction according to the reference feature coordinates.

[0301] Step 413 ′, calculate and obtain the characteristic height value between the upper boundary extreme value coordinates and the lower boundary extreme value coordinates to characterize the positional relationship between the maxillary model boundary and the mandibular model boundary.

[0302] Step 421', compare the characteristic height value with the height integrity criterion value representing the preset position condition, and judge whether the intraoral tissue area corresponding to the target tooth position in the oral three-dimensional model meets the preset integrity condition based on the comparison result.

[0303] Step 422, judging whether the intraoral tissue region corresponding to the target tooth position in the oral three-dimensional model satisfies a preset integrity condition based on the comparison result.

[0304] If not satisfied, jump to step 43, obtain the corresponding intraoral image data, extract the oral feature data corresponding to the intraoral tissue area in the intraoral image data, establish the pixel size relationship between the intraoral image data and the oral three-dimensional model data, and reconstruct the part representing the oral feature data on the oral three-dimensional model data to obtain the oral structure information.

[0305] Among them, the height training data includes the distance between the upper tissue extreme value coordinates on the three-dimensional training model corresponding to the intraoral tissue area located on one side of the first direction and the lower tissue extreme value coordinates on the three-dimensional training model corresponding to the intraoral tissue area located on the other side of the first direction.

[0306] Combination Figure 6 , Fig.17 and Fig.21 As shown, when the intraoral tissue area is determined to be the vestibular groove area, when the third training vestibular groove bottom point 1312U is the highest point on the upper vestibular groove of the three-dimensional training model data relative to the occlusal plane, and the sixth training vestibular groove bottom point 1341L is the lowest point on the lower vestibular groove of the three-dimensional training model data relative to the occlusal plane, the coordinates of the third training vestibular groove bottom point 1312U can be considered as the upper tissue extreme value coordinates, and the coordinates of the sixth training vestibular groove bottom point 1341L are the lower tissue extreme value coordinates, so as to obtain the distance between the two as the height training data. Of course, the above determination of the upper tissue extreme value coordinates and the lower tissue extreme value coordinates can only represent a special specific situation. For different three-dimensional training model data, the position pointed to by the above extreme value coordinates may have other situations.

[0307] The second embodiment and the first embodiment of this implementation are not necessarily independent of each other. The combination of the two can better establish the correspondence between the 3D training model data and the oral 3D model data, and provide a more specific and accurate solution for the calculation of the integrity condition. For example, in the second embodiment, the calculation process of the height training data can be consistent with the characteristic height value. Preferably, in the characteristic height value When the coordinate difference between the upper boundary extreme value point 101M and the lower boundary extreme value point 102M in the direction of the third coordinate axis Rz is configured, and the third coordinate axis Rz extends along the tooth midline, the height training data can be correspondingly configured as the distance value between the upper tissue extreme value coordinate and the lower tissue extreme value coordinate in the tooth midline extension direction, or the projection length value of the line between the two extreme value coordinate points in the tooth midline direction. Of course, if the calculation step of the characteristic height value is adjusted, the selection of the height training data can be adjusted accordingly.

[0308] The definitions of the average height data and the training height standard deviation may be similar to the average height data and the training distance standard deviation in the previous text, and the distribution probability coefficient may also be similar to the distribution probability coefficient corresponding to the distance training data in the previous text, which can be interpreted as the two differing only on the basis of data, one of which has a data basis pointing to the positional relationship between tooth position features and intraoral tissue features, and the other has a data basis pointing to the positional relationship between upper and lower intraoral tissue features, or between left and right intraoral tissue features, or between at least two intraoral tissues in other different positions.

[0309] For any specific example or derived technical solution in the second embodiment, further preferably, the distribution probability coefficient is 3.

[0310] For example, in an application scenario where the oral feature data includes vestibular sulcus height data, if the training data corresponding to all heights is calculated The average height data is , the standard deviation of training height is , then the distribution probability coefficient In a preferred embodiment of the invention, the high integrity criterion value , can be configured to at least satisfy:

[0311] .

[0312] Thus, the characteristic height values ​​can be compared , and the high integrity criterion value The relationship between the positional relationship and the preset positional condition corresponding to the positional relationship is achieved. It should be noted that since the characteristic height value and the high integrity criterion value are already data that characterize the integrity of the corresponding oral tissue, there is no need to count their number and set the allowable error quantity value. If the comparison result is that the characteristic height value is less than the high integrity criterion value, it can be directly judged that the oral three-dimensional model does not meet the preset integrity condition.

[0313] When the above two embodiments are combined, the following specific steps can be performed to determine whether the preset integrity condition is met: if the number of feature missing positions is greater than or equal to the allowable error number value, or the feature height value is less than the high integrity criterion value, then it is determined that the oral cavity three-dimensional model does not meet the preset integrity condition; if the feature missing position is less than the allowable error number value, and the feature height is greater than the high integrity criterion value, then it is determined that the oral cavity three-dimensional model meets the preset integrity condition. The feature extraction step after the preset integrity condition is met can be referred to the first embodiment, and will not be repeated here.

[0314] Based on any of the above implementations, Figure 4 and Fig.23 As shown, the present invention provides a first embodiment of a method for generating oral structure information, which refines step 43 and specifically provides steps 431 to 433. It can be understood that other steps except the refinement step will not be described in detail below. For example, the step 40 may specifically include steps 41 and 42. The first embodiment specifically includes the following steps.

[0315] Step 40, obtaining oral three-dimensional model data, and determining whether the intraoral tissue region specified in the oral three-dimensional model meets a preset integrity condition.

[0316] If not, jump to step 43, obtain the corresponding intraoral image data, extract the oral feature data corresponding to the intraoral tissue area in the intraoral image data, establish the pixel size relationship between the intraoral image data and the oral 3D model data, and reconstruct the part representing the oral feature data on the oral 3D model data to obtain the oral structure information. The step 43 specifically includes:

[0317] Step 431, calling the corresponding first neural network model and second neural network model according to the intraoral tissue area.

[0318] Step 432, input the intraoral image data into the first neural network model to extract the region of interest to obtain feature region image data.

[0319] Step 433, input the feature area image data into the second neural network model for feature recognition to obtain oral feature data corresponding to the intraoral tissue area.

[0320] In this way, the step of feature extraction in intraoral image data can be divided into two steps, reducing the amount of basic data for feature recognition, thereby speeding up the analysis of intraoral image data to obtain oral feature data. Using a neural network model to perform the above feature extraction steps can adapt to a variety of analysis scenarios, and is more accurate and more universal than simply using parameters such as chroma, RGB value or grayscale in the image.

[0321] In a specific example of the first embodiment, model training steps 71 to 74 for step 43 (that is, the above-mentioned steps 431 to 433) can be provided as model training steps. Specifically, the steps 71 to 74 can be set at any position before step 431, and the present invention is not limited to this. Hereinafter, the present invention will be described by taking the specific example in which the pre-step is set before step 41 as an example. It can be understood that other steps except the pre-step will not be described in detail below. The specific example may include the following steps.

[0322] Step 71: Obtain at least two sets of training image data and region of interest labels corresponding to the training image data.

[0323] Step 72, calling a preset convolutional neural network model, using the training image data and the region of interest mark as model input for iterative training, and obtaining first training parameters and a corresponding first neural network model.

[0324] Step 73, obtaining at least two sets of training image data, and tooth position markings and vestibular groove bottom markings corresponding to the training image data.

[0325] Step 74, calling a preset convolutional neural network model, using the training image data, tooth position marks and vestibular groove bottom marks as model inputs for iterative training, and obtaining second training parameters and a corresponding second neural network model.

[0326] Step 40, obtaining oral three-dimensional model data, and determining whether the intraoral tissue region specified in the oral three-dimensional model meets a preset integrity condition.

[0327] If not, jump to step 43, obtain the corresponding intraoral image data, extract the oral feature data corresponding to the intraoral tissue area in the intraoral image data, establish the pixel size relationship between the intraoral image data and the oral 3D model data, and reconstruct the part representing the oral feature data on the oral 3D model data to obtain the oral structure information. The step 43 specifically includes:

[0328] Step 431, calling the corresponding first neural network model and second neural network model according to the intraoral tissue area.

[0329] Step 432, input the intraoral image data into the first neural network model to extract the region of interest to obtain feature region image data.

[0330] Step 433, input the feature area image data into the second neural network model for feature recognition to obtain oral feature data corresponding to the intraoral tissue area.

[0331] The extension range of the region of interest marker in the corresponding training image at least covers the upper lip, lower lip and dentition in the training image.

[0332] Using convolutional neural networks to build the first neural network model and the second neural network model can better adapt to the use of local feature analysis. On the one hand, its sparse connection, weight sharing and downsampling characteristics can be used to reduce the amount of data in the calculation process; on the other hand, its translation invariance can be used to ensure the accuracy of feature recognition.

[0333] In this embodiment, the extension range of the region of interest mark is limited, and the location for feature extraction is determined to be in the area where the upper lip, lower lip and teeth are located. This can avoid the influence of human facial features on the feature extraction process of the intraoral image, and also avoid the extraction of unnecessary features that leads to an increase in the amount of computing data.

[0334] Based on any of the above implementations, Fig.24 As shown, the present invention provides a second embodiment of a method for generating oral structure information, which refines step 43 and specifically provides steps 434 to 435. It can be understood that other steps except the refinement step will not be described in detail below. For example, step 40 may specifically include steps 41 and 42. The second embodiment specifically includes the following steps.

[0335] Step 40, obtaining oral three-dimensional model data, and determining whether the intraoral tissue region specified in the oral three-dimensional model meets a preset integrity condition.

[0336] If not, jump to step 43, obtain the corresponding intraoral image data, extract the oral feature data corresponding to the intraoral tissue area in the intraoral image data, establish the pixel size relationship between the intraoral image data and the oral 3D model data, and reconstruct the part representing the oral feature data on the oral 3D model data to obtain the oral structure information. The step 43 specifically includes:

[0337] Step 434, determine at least one target tooth position corresponding to each other in the intraoral image and the oral three-dimensional model as a relative reference tooth position, and calculate the size data of the crown of the relative reference tooth position in at least one same direction in the intraoral image and the oral three-dimensional model, and obtain the reference pixel size data and the reference physical size data respectively.

[0338] Step 435 , fitting a size mapping factor based on the reference pixel size data and the reference physical size data to characterize the pixel size relationship.

[0339] Two points need to be explained. First, the second embodiment can be combined with the first embodiment and the third embodiment to form a more complete step 43. Of course, the three embodiments can also achieve corresponding technical effects respectively. Fig.25 Give an example description of the details. Fig.25 The Z2 and Z3 parts in the figure are corresponding local schematic diagrams, where the Z2 part points to Fig.21 , is the part of the intraoral tissue structure focused on the central incisor position of the oral 3D model, and the Z3 part is the part of the intraoral tissue structure corresponding to the intraoral image at the central incisor position, citing Fig.25 In order to specifically illustrate the steps of establishing the pixel size relationship, the present invention is not limited in terms of the selection of the target tooth position and the like.

[0340] Combination Fig.21 and Fig.25 As shown, for example, if the maxillary left central incisor tooth position in Z2 and Z3 is selected as the relative reference tooth position, the pixel size relationship can be established with the size data of the tooth position in the same direction. For example, the length, width or area of ​​the crown lip surface can be used as the size data, thereby obtaining the reference pixel size data corresponding to the intraoral image and the reference physical size data corresponding to the oral three-dimensional model.

[0341] Specifically, the unit of the reference pixel size data is pixel, or can be interpreted as the number of pixels, and the unit of the reference physical size is millimeter, or can be interpreted as the length of the size data. In this way, a mapping relationship between a plane image and a three-dimensional image can be established, which is convenient for subsequent reconstruction of oral features.

[0342] Preferably, the at least one same direction includes the crown width direction. In this way, the establishment of pixel size relationship can be avoided to be affected by the inconsistency of the anterior and posterior crown lengths and the crown labial surface area caused by special circumstances such as gum atrophy, and the crown width with a relatively stable value can be used as the data basis for establishing the pixel size relationship.

[0343] Preferably, the size mapping factor is the quotient of the reference physical size data and the reference pixel size data. In this way, the physical size represented by a single pixel in the intraoral image can be calculated, that is, the length value (in millimeters) of a single pixel corresponding to the oral 3D model. This "scale"-like data is used to ensure the stability of the feature correspondence during the reconstruction process.

[0344] On this basis, the present invention provides a specific example based on the above second embodiment, which specifically refines the process of calculating the reference pixel size data and the reference physical size data, and completes the establishment of the mapping relationship between the two by selecting reference feature points at the corresponding tooth positions on the intraoral image and the oral three-dimensional model. Fig.26 or Fig. 27 As shown, this specific example is basically consistent with other steps in the second embodiment described above, but the following detailed steps are provided for step 434.

[0345] Step 81, determining a first reference feature point and a second reference feature point relative to a reference tooth position on the intraoral image, and calculating and obtaining reference pixel size data based on the number of pixels between the first reference feature point and the second reference feature point.

[0346] Step 82, determining a third reference feature point and a fourth reference feature point relative to the reference tooth position on the three-dimensional oral model, and calculating and obtaining reference physical dimension data based on the Euclidean distance between the third reference feature point and the fourth reference feature point.

[0347] Wherein, the first reference feature point and the second reference feature point are located on both sides of the long axis of the tooth body of the relative reference tooth position, and the distance between the first reference feature point and the corresponding relative reference tooth position crown incision is equal to the distance between the second reference feature point and the corresponding relative reference tooth position crown incision. The third reference feature point and the fourth reference feature point are located on both sides of the long axis of the tooth body of the relative reference tooth position, and the distance between the third reference feature point and the corresponding relative reference tooth position crown incision is equal to the distance between the fourth reference feature point and the corresponding relative reference tooth position crown incision.

[0348] Combination Fig.25 As shown, on the intraoral image, if the relative reference tooth position is determined to be the maxillary left central incisor tooth position, the first reference feature point and the second reference feature point can be points located on the mesial boundary and the distal boundary on both sides of the long axis 1121r of the maxillary left central incisor tooth body. If the relative reference tooth position is determined to be the maxillary right central incisor tooth position, the first reference feature point and the second reference feature point can be points located on the mesial boundary and the distal boundary on both sides of the long axis 1111s of the maxillary right central incisor tooth body. If the relative reference tooth position is determined to be the maxillary left central incisor tooth position and the maxillary right central incisor tooth position, the first reference feature point and the second reference feature point can also be points located on the distal boundary of the maxillary left central incisor tooth position and the distal boundary of the maxillary right central incisor tooth position on both sides of the long axis of the tooth bodies. In addition, the distances of the above two reference feature points relative to the incisal end of the crown are equal, which can ensure that they accurately reflect the width characteristics of the crown.

[0349] Correspondingly, in the three-dimensional model of the oral cavity, the long axis 1121r' of the maxillary left central incisor and the long axis 1111r' of the maxillary right central incisor are also included. Based on a technical solution similar to the first reference feature point and the second reference feature point, at least the above-mentioned three selection solutions for the third reference feature point and the fourth reference feature point can be obtained.

[0350] Under the technical route of this specific example, Fig.24 and Fig.26 , the present invention further provides a first specific example of the second embodiment. The first specific example mainly refines step 434, adds a crown integrity judgment step 801 to the step, and refines step 802A executed after judging that the crown is intact, as well as step 811A and step 812A belonging to step 81, so as to achieve the determination of the relative reference tooth position and the calculation of the reference pixel size. It can be understood that for steps other than the refinement step, the following description will not be expanded. For example, step 40 may specifically include step 41 and step 42. Similarly, since steps 801 to 82 belong to step 434, the description of step 434 itself will be omitted below; since steps 811A and 812A belong to step 81, the description of step 81 itself will be omitted below. The first specific example specifically includes the following steps.

[0351] Step 40, obtaining oral three-dimensional model data, and determining whether the intraoral tissue region specified in the oral three-dimensional model meets a preset integrity condition.

[0352] If not, jump to step 43, obtain the corresponding intraoral image data, extract the oral feature data corresponding to the intraoral tissue area in the intraoral image data, establish the pixel size relationship between the intraoral image data and the oral 3D model data, and reconstruct the part representing the oral feature data on the oral 3D model data to obtain the oral structure information. The step 43 specifically includes:

[0353] Step 430, obtaining corresponding intraoral image data.

[0354] Step 801, determining whether the maxillary left central incisor tooth position and the maxillary right central incisor tooth position in the intraoral image and the oral three-dimensional model both contain complete tooth crowns.

[0355] If yes, jump to step 802A to determine the corresponding maxillary left central incisor tooth positions and maxillary right central incisor tooth positions in the intraoral image and the oral three-dimensional model as relative reference tooth positions.

[0356] Step 811A, determine the first reference feature point at the distal boundary of the crown of the maxillary left central incisor tooth position in the intraoral image, and determine the second reference feature point at the distal boundary of the crown of the maxillary right central incisor tooth position in the intraoral image.

[0357] Step 812A, calculating the number of pixels between the first reference feature point and the second reference feature point, and using half of the number of pixels as reference pixel size data.

[0358] Step 82, determining a third reference feature point and a fourth reference feature point relative to the reference tooth position on the three-dimensional oral model, and calculating and obtaining reference physical dimension data based on the Euclidean distance between the third reference feature point and the fourth reference feature point.

[0359] Step 435 , fitting a size mapping factor based on the reference pixel size data and the reference physical size data to characterize the pixel size relationship.

[0360] In this way, the specific example of the second embodiment can cope with special situations such as missing teeth, and adaptively change the strategy for establishing pixel size relationships, thereby having a wider range of application scenarios. It can be understood that the specific example of the above preferred embodiment uses the central incisor tooth position as a reference to obtain clearer data, but the present invention does not exclude the use of other tooth positions for pixel size relationship fitting technical solutions.

[0361] Whether the crown is complete can be determined based on the fitting of the convex hull contour, for example, determining whether the area or distribution of the convex hull contour is uniform, whether the convex hull contour can be fitted, etc., which will not be elaborated here.

[0362] When judging that the crowns are intact, the overall width of the crowns of the two maxillary central incisors can be used as the basis for fitting the pixel size relationship, which has higher accuracy. Fig.25 As shown, in the intraoral image, the pixel point 1121s at the distal boundary of the maxillary left central incisor can be determined as the first reference feature point, and the pixel point 1111s at the distal boundary of the maxillary right central incisor can be determined as the second reference feature point, and the number of pixels between the above two pixel points is calculated, which is directly used as the reference pixel size data. Preferably, half of it is calculated as the reference pixel size data, which can reflect the average pixel situation of the central incisor width.

[0363] Correspondingly, step 82 may also include steps corresponding to step 811A and step 812A. The specific step description is not repeated here. Fig.25It can be seen that in the oral three-dimensional model, the distal boundary space point 1121s' of the maxillary left central incisor can be determined as the third reference feature point, and the distal boundary space point 1111s' of the maxillary right central incisor can be determined as the fourth reference feature point, and the Euclidean distance between the above two pixel points can be calculated.

[0364] Correspondingly, if Fig.24 and Fig. 27 , the present invention further provides a second specific example of the second embodiment. The second specific example provides a further judgment step 802B after judging that the crown is incomplete, as well as step 811B and step 8112B belonging to step 81. It can be understood that other steps except the refinement step will not be described in detail below. For example, step 40 may specifically include step 41 and step 42. Similarly, since steps 801 to 82 belong to step 434, the description of step 434 itself will be omitted below; since steps 811B and 812B belong to step 81, the description of step 81 itself will be omitted below. The second specific example specifically includes the following steps.

[0365] Step 40, obtaining oral three-dimensional model data, and determining whether the intraoral tissue region specified in the oral three-dimensional model meets a preset integrity condition.

[0366] If not, jump to step 43, obtain the corresponding intraoral image data, extract the oral feature data corresponding to the intraoral tissue area in the intraoral image data, establish the pixel size relationship between the intraoral image data and the oral 3D model data, and reconstruct the part representing the oral feature data on the oral 3D model data to obtain the oral structure information. The step 43 specifically includes:

[0367] Step 430, obtaining corresponding intraoral image data.

[0368] Step 801, determining whether the maxillary left central incisor tooth position and the maxillary right central incisor tooth position in the intraoral image and the oral three-dimensional model both contain complete tooth crowns.

[0369] If not, (step 802B) and only the maxillary left central incisor tooth position in the intraoral image and the oral three-dimensional model contains a complete crown, the maxillary left central incisor tooth position corresponding to each other in the intraoral image and the oral three-dimensional model is determined as the relative reference tooth position.

[0370] Step 811B, determine the first reference feature point at the distal boundary of the crown of the maxillary left central incisor tooth position in the intraoral image, and determine the second reference feature point at the mesial boundary of the crown of the maxillary left central incisor tooth position in the intraoral image.

[0371] Step 812B, calculating the number of pixels between the first reference feature point and the second reference feature point, and using the number of pixels as reference pixel size data.

[0372] Step 82, determining a third reference feature point and a fourth reference feature point relative to the reference tooth position on the three-dimensional oral model, and calculating and obtaining reference physical dimension data based on the Euclidean distance between the third reference feature point and the fourth reference feature point.

[0373] Step 435 , fitting a size mapping factor based on the reference pixel size data and the reference physical size data to characterize the pixel size relationship.

[0374] In this specific example, the crown loss of the central incisor is also determined, which has a similar technical effect to the previous specific example. The two can also be combined to form a complete specific example included in the second embodiment. In addition, although the text description of the present invention is based on the fact that the maxillary left central incisor contains a complete crown, it can be understood that the technical solution provided by the present invention can be implemented for different actual situations to achieve the corresponding technical effects.

[0375] In this case, combined with Fig.25 As shown, in the intraoral image, the distal boundary pixel point 1121s of the maxillary left central incisor can be determined as the first reference feature point, and the mesial boundary pixel point Cs of the maxillary left central incisor can be determined as the second reference feature point, and the number of pixels between the above two pixel points is calculated.

[0376] Correspondingly, step 82 may also include steps corresponding to step 811B and step 812B. The specific step description is not repeated here. Fig.25 It can be seen that in the three-dimensional model of the oral cavity, the distal boundary space point 1121s' of the maxillary left central incisor can be determined as the third reference feature point, and the mesial boundary space point Cs' of the maxillary left central incisor can be determined as the fourth reference feature point, and the Euclidean distance between the above two pixel points can be calculated.

[0377] In other cases, the above-mentioned maxillary left central incisor mesial boundary pixel point Cs can also be interpreted as the maxillary right central incisor mesial boundary pixel point, and the above-mentioned maxillary left central incisor mesial boundary space point Cs' can also be interpreted as the maxillary right central incisor mesial boundary space point. Based on this, the number of pixels between the maxillary right central incisor distal boundary pixel point 1111s and the mesial boundary pixel point, as well as the Euclidean distance between the maxillary right central incisor distal boundary space point 1111s' and the mesial boundary space point can also be calculated.

[0378] For any specific example or derived technical solution in the second embodiment, it is further preferred that the distance from the first reference feature point to the gingival end of the corresponding incisor tooth position is three times the distance from the first reference feature point to the incisal end of the crown of the corresponding incisor tooth position. In other words, the distance from the first reference feature point to the incisal end of the crown can be one-fourth of the length of the crown, and the width value extracted in this way is used as the basis for establishing the pixel size relationship, which has stronger stability. Based on this, those skilled in the art can understand that the second reference feature point, the third reference feature point and the fourth reference feature point can also be configured to have the above-mentioned position features.

[0379] Based on any of the above implementations, Figure 4 and Fig.28 As shown, the present invention provides a third embodiment of a method for generating oral structure information, which refines step 43 and specifically provides steps 436 to 437. It can be understood that other steps except the refinement step will not be described in detail below. For example, step 40 may specifically include step 41 and step 42. The third embodiment specifically includes the following steps.

[0380] Step 40, obtaining oral three-dimensional model data, and determining whether the intraoral tissue region specified in the oral three-dimensional model meets a preset integrity condition.

[0381] If not, jump to step 43, obtain the corresponding intraoral image data, extract the oral feature data corresponding to the intraoral tissue area in the intraoral image data, establish the pixel size relationship between the intraoral image data and the oral 3D model data, and reconstruct the part representing the oral feature data on the oral 3D model data to obtain the oral structure information. The step 43 specifically includes:

[0382] Step 436, determining at least two intraoral tissue feature points in the intraoral image data according to the oral cavity feature data.

[0383] Step 437, based on the intraoral tissue feature points, at the feature missing position in the oral three-dimensional model, the tissue space feature points corresponding to the intraoral tissue feature points are fitted according to the pixel size relationship, and a part representing the oral feature data is formed on the oral three-dimensional model data.

[0384] Based on this, at least when the oral feature data includes vestibule height data, dental arch width data, lip frenulum width data or maxillofacial protrusion amplitude data, the most representative tissue space feature points can be reconstructed at the feature missing position of the oral 3D model. For cases where the oral feature data includes dental arch curvature data, it is necessary to determine multiple oral tissue feature points to fit the overall curvature curve of the dental arch on the oral 3D model.

[0385] In one application scenario, preferably, the oral tissue area includes the vestibule groove, the oral feature data includes the vestibule groove height data, and the oral tissue feature points include the upper groove bottom feature points and the lower groove bottom feature points. In this way, the most typical feature points can be extracted for reference by medical workers.

[0386] In this application scenario, continue as Fig.28 As shown, the above step 437 may further include the following steps.

[0387] Step 4371, calculate the distance between the upper groove bottom feature point and the midpoint of the gingival margin of the corresponding tooth position to obtain the upper groove bottom spacing parameter, and calculate the corresponding upper groove bottom mapping parameter based on the upper groove bottom spacing parameter and the size mapping factor that characterizes the pixel size relationship.

[0388] Step 4372, based on the reference feature coordinates of the target tooth position corresponding to the upper groove bottom feature point and the upper groove bottom mapping parameters, the upper spatial feature point representing the upper groove bottom position is fitted.

[0389] Step 4373, calculate the distance between the lower groove bottom feature point and the midpoint of the gingival margin of the corresponding tooth position to obtain the lower groove bottom spacing parameter, and calculate the corresponding lower groove bottom mapping parameter based on the lower groove bottom spacing parameter and the size mapping factor.

[0390] Step 4374, based on the reference characteristic coordinates of the target tooth position corresponding to the lower groove bottom characteristic point and the lower groove bottom mapping parameters, the lower spatial characteristic points representing the lower groove bottom position are fitted.

[0391] In this way, the midpoint of the gingival margin, which is also used as a reference point in the oral 3D model, can be used to fit the mapping relationship between the upper and lower groove feature points, thereby maintaining the consistency between the intraoral image and the oral 3D model. Figure 6 As shown, at least the coordinate point 13A of the bottom of the first vestibule groove can be fitted on the three-dimensional model of the oral cavity as an upper spatial feature point.

[0392] Of course, the present invention is not limited to the above application scenarios. In another application scenario, preferably, the intraoral tissue area includes the vestibule groove, or the tooth root ridge, or both the vestibule groove and the tooth root ridge, the oral feature data includes the dental arch width data, and the intraoral tissue feature points include the left groove bottom feature points and the right groove bottom feature points. In this application scenario, the establishment of the mapping relationship and the fitting of the feature points are similar to those in the previous application scenario, and will not be repeated here.

[0393] Of course, the embodiments provided by the present invention are not limited to fitting only some spatial feature points in the oral three-dimensional model, and of course all spatial feature points can be fitted, and even a spatial distribution curve or a tissue spatial distribution surface can be obtained by fitting. Based on this, in a specific example of the third embodiment above, the steps 436 and 437 can be further configured as the following steps 436' and 437'.

[0394] Step 436 ′: in the intraoral image, determine all intraoral tissue feature points according to the oral cavity feature data.

[0395] Step 437', according to the intraoral tissue feature points, at the feature missing position in the oral three-dimensional model, the tissue space feature points corresponding to the intraoral tissue feature points are fitted according to the pixel size relationship, and the tissue space distribution curve or tissue space distribution surface is fitted according to the tissue space feature points, and this is used as the part representing the oral feature data on the oral three-dimensional model data.

[0396] Likewise, combining Figure 6 As shown, the above preferred technical solution can be applied to achieve the fitting of the first tissue space distribution curve 13a or the second tissue space distribution curve 13b, and can also achieve the fitting of the first tissue space distribution surface Sa. In particular, the fitting of the tissue space distribution curve and the tissue space distribution surface, or the coordinate value of the tissue space feature point in the direction of the second coordinate axis Ry (or, the direction perpendicular to the paper surface), can be determined based on the grayscale value, brightness value of the corresponding point in the intraoral image, or the actual depth on the intraoral side view image.

[0397] like Fig.29 As shown, another embodiment of the present invention provides a method for generating oral structure information. The application program or instruction corresponding to the method can be installed in the above storage medium and / or the above oral structure information generating system 300 to achieve the technical effect of generating oral structure information. The method for generating oral structure information can specifically include the following steps.

[0398] Step 91, obtaining oral cavity three-dimensional model data and corresponding intraoral image data.

[0399] Step 92, extract the oral feature data corresponding to the target intraoral tissue area in the oral three-dimensional model from the intraoral image data, establish a pixel size relationship between the intraoral image data and the oral three-dimensional model data, and reconstruct the part representing the oral feature data on the oral three-dimensional model data to obtain oral structure information.

[0400] The above-mentioned embodiment is compared with the present invention in Figure 4 , Figure 5 or Fig.19With regard to the implementation method provided in the embodiment, as well as other embodiments or specific examples obtained by combining and appending the steps of the two, the step of judging the integrity of the oral three-dimensional model is eliminated, and the intraoral image data is directly used for feature extraction, and the extracted features are reconstructed on the oral three-dimensional model, thereby avoiding the processing of complex point cloud data on the oral three-dimensional model, and being able to display more intuitive feature points, curves or surfaces on the oral three-dimensional model.

[0401] Those skilled in the art can understand that each step in this embodiment can be replaced by the description or limitation of the steps in other embodiments, examples or specific examples provided above. In particular, the extraction of oral feature data, the establishment of pixel size relationship and the reconstruction of the oral feature data part are not described in detail in the present invention.

[0402] Based on any of the above embodiments, the oral structure information generated by the present invention can be interpreted as an oral three-dimensional model and its data including the "part representing the oral feature data", or it can be interpreted as referring only to the "part representing the oral feature data" (that is, at least one of the feature points, curves or surfaces).

[0403] Of course, in some embodiments, the oral structure information can also be interpreted as a kind of data, preferably oral feature data after pixel size relationship mapping. Specifically, it can be the value of the vestibule sulcus height on the oral 3D model, the value of the dental arch width on the oral 3D model, the value of the lip frenulum width on the oral 3D model, the value of the dental arch curvature on the oral 3D model, or the value of the maxillofacial protrusion amplitude / curvature on the oral 3D model.

[0404] It should be noted that although a first direction D1 is marked in some drawings of the specification, it does not mean that the corresponding implementation cannot adopt a technical solution of setting different first directions for different tooth positions. Figure 6 , Fig.15 , Fig.17 , Fig.25 In the figure, the first direction D1 is only used as an example, which can be interpreted as an example of the first direction of the central incisor corresponding to the three-dimensional oral model in the positive viewing angle direction. For the first direction of the three-dimensional oral model in other viewing angle directions or for other tooth positions, Fig.12 and Fig.13 It can be deduced.

[0405] In addition, the above-mentioned multiple embodiments corresponding to a certain implementation mode can be combined as a whole or in part, for example Figure 4The three embodiments and their specific examples corresponding to the provided implementation modes can be combined; the above-mentioned multiple embodiments corresponding to a certain implementation mode can be combined into another implementation mode or yet another implementation mode, for example Figure 4 The provided embodiments can be used with Figure 5 Another embodiment provided is combined as a whole, for example Figure 4 and / or Figure 5 The provided embodiments can be used with Fig.19 Another embodiment provided is combined as a whole to set higher integrity condition judgment requirements, for example corresponding to Figure 4 The first to third embodiments corresponding to the provided implementation modes can be Fig.19 The provided further implementation manner is combined to enrich the content of the further implementation manner and achieve corresponding effects.

[0406] In summary, the oral structure information generation method provided by the present invention analyzes the relative position relationship between the tooth position features and the model boundary features, and when it is determined that the oral three-dimensional model does not meet the preset conditions, calls the corresponding intraoral image data to complete the features, thereby obtaining complete oral structure information based on the oral three-dimensional model data, thereby reducing the requirements for extracting the oral three-dimensional model, and simplifies the steps and operation logic on the basis of obtaining high-precision oral structure features; the composite process between the intraoral image data and the oral three-dimensional model data is mainly to reconstruct the features on the oral three-dimensional model according to the established pixel size relationship, and the output oral structure information is based on three-dimensional data, has strong intuitiveness and accuracy, is convenient for medical workers to conduct further analysis, and is also convenient for patients or other relevant personnel to consult.

[0407] It should be understood that although this specification is described according to implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each implementation mode may also be appropriately combined to form other implementation modes that can be understood by those skilled in the art.

[0408] The series of detailed descriptions listed above are only specific descriptions of feasible implementation methods of the present invention. They are not intended to limit the scope of protection of the present invention. Any equivalent implementation methods or changes that do not deviate from the technical spirit of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for generating oral structure information, characterized in that: include: Acquire oral three-dimensional model data and determine whether the intraoral tissue area specified in the oral three-dimensional model meets the preset integrity condition; If not, then obtain the corresponding intraoral image data, extract the oral feature data corresponding to the intraoral tissue area in the intraoral image data, establish a pixel size relationship between the intraoral image data and the oral 3D model data, and reconstruct the part representing the oral feature data on the oral 3D model data to obtain the oral structure information; The “reconstructing the part representing the oral characteristic data on the oral three-dimensional model data to obtain the oral structure information” specifically includes: In the intraoral image data, determining at least two intraoral tissue feature points according to the oral cavity feature data; According to the intraoral tissue feature points, at the feature missing position in the oral three-dimensional model, tissue space feature points corresponding to the intraoral tissue feature points are fitted according to the pixel size relationship, and a part representing the oral feature data is formed on the oral three-dimensional model data; The "obtaining oral three-dimensional model data and determining whether the intraoral tissue area specified in the oral three-dimensional model meets the preset integrity condition" specifically includes: Acquire oral three-dimensional model data, and calculate the positional relationship between the target tooth position and the model boundary and / or the positional relationship between the maxillary model boundary and the mandibular model boundary according to the target tooth position features in the oral three-dimensional model data and the model boundary features corresponding to the target tooth position features; Comparing the positional relationship with a preset position condition corresponding to the positional relationship, and determining whether the intraoral tissue region corresponding to the target tooth position in the oral three-dimensional model satisfies a preset integrity condition; The “establishing a pixel size relationship between the intraoral image data and the oral three-dimensional model data” specifically includes: Determine at least one target tooth position corresponding to each other in the intraoral image and the oral three-dimensional model as a relative reference tooth position, and calculate the size data of the crown of the relative reference tooth position in at least one same direction in the intraoral image and the oral three-dimensional model, and obtain reference pixel size data and reference physical size data respectively; A size mapping factor is obtained by fitting according to the reference pixel size data and the reference physical size data to characterize the pixel size relationship.

2. The method for generating oral structure information according to claim 1, characterized in that: The “calculating the positional relationship between the target tooth position and the model boundary according to the target tooth position features in the oral three-dimensional model data and the model boundary features corresponding to the target tooth position features” specifically includes: Determine at least one tooth position on the oral three-dimensional model as a target tooth position, and calculate the reference feature coordinates of the target tooth position based on a preset feature recognition rule to characterize the features of the target tooth position; According to the reference feature coordinates, determining boundary feature coordinates corresponding to the reference feature coordinates on the oral cavity three-dimensional model along a first direction to characterize the model boundary features; Calculating a characteristic distance value between the reference characteristic coordinates and the boundary characteristic coordinates to characterize a positional relationship between the target tooth position and the model boundary; The "comparing the positional relationship with the preset positional condition corresponding to the positional relationship to determine whether the intraoral tissue region corresponding to the target tooth position in the oral three-dimensional model satisfies the preset integrity condition" specifically includes: The characteristic distance value is compared with the distance integrity criterion value characterizing the preset position condition, and based on the comparison result, it is determined whether the intraoral tissue area corresponding to the target tooth position in the oral three-dimensional model meets the preset integrity condition.

3. The method for generating oral structure information according to claim 2, characterized in that: The reference feature coordinates are located in a preset model coordinate system, and the origin of the model coordinate system is located in the occlusal plane; the "calculating the reference feature coordinates of the target tooth position based on a preset feature recognition rule" specifically includes: Determine all spatial points on the target tooth crown located on the labial side to form a set of spatial points; Selecting the spatial point with the smallest coordinate value in the set of spatial points as the pole to establish a polar coordinate system, and arranging the other spatial points in ascending order according to the polar angle and polar diameter to form a feature traversal sequence; Extracting the coordinates of the first two spatial points in the feature traversal sequence, taking the pole as the starting point, sequentially calculating a first polar coordinate vector and a second polar coordinate vector, and calculating the cross product of the first polar coordinate vector and the second polar coordinate vector, and determining whether the cross product is less than 0; If yes, then update the starting point to be a first spatial point corresponding to the first polar coordinate vector; if no, then update the starting point to be a second spatial point corresponding to the second polar coordinate vector; Traversing all other spatial points located after the second spatial point in the feature traversal sequence, selectively updating the starting point according to the polar coordinate vector formed by the spatial point and the starting point, and the cross product between the polar coordinate vectors, and determining at least two spatial points that meet the determination condition as convex hull reference points; According to the convex hull reference point and the pole, the convex hull contour data of the target tooth position is fitted, and according to the convex hull contour data, the reference feature coordinates of the target tooth position are calculated.

4. The method for generating oral structure information according to claim 3, characterized in that: The "traversing all other spatial points after the second spatial point in the feature traversal sequence, updating the starting point according to the polar coordinate vector formed by the spatial point and the starting point, and the cross product between the polar coordinate vectors, and determining at least two spatial points that meet the determination condition as convex hull reference points" specifically includes: Extracting a third spatial point located after the second spatial point in the feature traversal sequence, sequentially calculating a reference polar coordinate vector between a reference spatial point other than the starting point in the first spatial point and the second spatial point and the starting point, and a third polar coordinate vector between the third spatial point and the starting point, and calculating a cross product of the reference polar coordinate vector and the third polar coordinate vector, and determining whether the cross product is less than 0; If yes, then update the starting point to the reference space point, and determine the reference space point as the convex hull reference point; if no, do not update the starting point, delete the reference space point, use the third space point as the new reference space point, and selectively update the starting point and determine the convex hull reference point based on the new reference polar coordinate vector and the cross product of the polar coordinate vector formed by the starting point and the next space point of the new reference space point; The iteration is repeated until all spatial points in the feature traversal sequence are judged, and at least two convex hull reference points are obtained.

5. The method for generating oral structure information according to claim 2, characterized in that: The "calculating the reference feature coordinates of the target tooth position based on the preset feature recognition rules" specifically includes: The coordinates of the midpoint of the gingival margin of the target tooth position are determined and used as the reference feature coordinates of the target tooth position.

6. The method for generating oral structure information according to claim 2, characterized in that: The “determining, according to the reference feature coordinates, boundary feature coordinates corresponding to the reference feature coordinates on the oral three-dimensional model along a first direction to characterize the model boundary features” specifically includes: Taking the reference feature coordinates as a starting point, a reference extension line is drawn along the first direction in a direction away from the target tooth position, and continuously analyzing the formation of projection endpoints of the reference extension ends on the reference extension line other than the reference feature coordinates on the jaw surface of the oral three-dimensional model; When the reference extension end no longer forms a projection endpoint with the maxillofacial surface of the oral three-dimensional model, the last formed projection endpoint coordinates are determined as the boundary feature coordinates.

7. The method for generating oral structure information according to claim 6, characterized in that: The intraoral tissue area includes the vestibule groove, the oral feature data includes the vestibule groove height data, the first direction is the projection of the extension direction of the long axis of the tooth on the maxillary surface of the oral three-dimensional model, and the target tooth position includes the maxillary incisor tooth position and the mandibular incisor tooth position.

8. The method for generating oral structure information according to claim 6, characterized in that: The intraoral tissue area includes the vestibule groove and / or the root ridge, the oral feature data includes dental arch width data, the first direction is the projection of the extension direction of the long axis of the tooth on the maxillary surface of the oral three-dimensional model, and the target tooth position includes the corresponding left molar tooth position of the first maxillary surface and the right molar tooth position of the first maxillary surface; wherein, the first maxillary surface is the maxilla and / or mandible.

9. The method for generating oral structure information according to claim 2, characterized in that: The intraoral tissue area includes the lip frenulum, the oral characteristic data includes the lip frenulum width data, and the "determining at least one tooth position on the oral three-dimensional model as a target tooth position, and calculating the reference characteristic coordinates of the target tooth position based on a preset feature recognition rule to characterize the target tooth position feature" specifically includes: Determine the left central incisor tooth position of the first maxillary face and the right central incisor tooth position of the first maxillary face on the three-dimensional oral model as the target tooth position, and calculate the reference feature coordinates of the target tooth position and a plurality of relative feature coordinates on the target tooth position located between the two reference feature coordinates based on a preset feature recognition rule to characterize the target tooth position features; wherein the first maxillary face is the maxilla and / or the mandible; The “determining, according to the reference feature coordinates, boundary feature coordinates corresponding to the reference feature coordinates on the oral three-dimensional model along a first direction to characterize the model boundary features” specifically includes: Taking the reference feature coordinates and the relative feature coordinates as starting points, respectively, a reference extension line is drawn along a first direction away from the target tooth position, and the formation of projection endpoints of the reference extension ends on the reference extension line other than the reference feature coordinates or the relative feature coordinates on the three-dimensional model of the oral cavity is continuously analyzed; wherein the first direction is the projection of the extension direction of the long axis of the tooth on the jaw surface of the three-dimensional model of the oral cavity; When the reference extension end no longer forms the projection endpoint with the oral three-dimensional model, the last formed projection endpoint coordinates are determined as the boundary feature coordinates.

10. The method for generating oral structure information according to claim 2, characterized in that: Before the step of "comparing the characteristic distance value with the distance integrity criterion value representing the preset position condition, and judging whether the intraoral tissue area corresponding to the target tooth position in the oral three-dimensional model meets the preset integrity condition according to the comparison result", the step includes: Acquire at least two sets of three-dimensional training model data, and at least two sets of distance training data corresponding to the three-dimensional training model data; wherein the distance training data includes the distance between the conditional training coordinates corresponding to the reference feature coordinates on the three-dimensional training model and the target tissue coordinates corresponding to the intraoral tissue area on the three-dimensional training model; the line connecting the conditional training coordinates and the target tissue coordinates extends along the first direction; The average distance data and the training distance standard deviation of the distance training data are calculated, and the distance integrity criterion value is calculated according to the difference between the average distance data and the product of the training distance standard deviation and a preset distribution probability coefficient.

11. The method for generating oral structure information according to claim 10, characterized in that: The intraoral tissue area includes the vestibule groove, the oral feature data includes the vestibule groove height data, the target tooth position includes the maxillary first incisor tooth position and the mandibular second incisor tooth position, the distance training data includes a first groove bottom distance parameter corresponding to the maxillary first incisor tooth position, and a second groove bottom distance parameter corresponding to the mandibular second incisor tooth position; the "calculating the average distance data and the training distance standard deviation of the distance training data" specifically includes: Calculating the first ditch bottom average distance value of all first ditch bottom distance parameters in all distance training data, and the second ditch bottom average distance value of all second ditch bottom distance parameters in all distance training data, to obtain the average distance data; The first groove bottom distance standard deviation of all first groove bottom distance parameters in all distance training data and the second groove bottom distance standard deviation of all second groove bottom distance parameters in all distance training data are calculated to obtain the training distance standard deviation.

12. The method for generating oral structure information according to claim 11, characterized in that: The distance training data includes groove bottom distance parameters corresponding to the maxillary left lateral incisor tooth position, maxillary left central incisor tooth position, maxillary right central incisor tooth position, maxillary right lateral incisor tooth position, mandibular left lateral incisor tooth position, mandibular left central incisor tooth position, mandibular right central incisor tooth position and mandibular right lateral incisor tooth position respectively; the groove bottom distance parameters represent the distance between the conditional training coordinates of the midpoint of the gingival margin of the corresponding incisor tooth position and the target tissue coordinates of the groove bottom feature point corresponding to the midpoint of the gingival margin.

13. The method for generating oral structure information according to claim 10, characterized in that: The intraoral tissue area includes the vestibule groove and / or the root ridge, the oral characteristic data includes the dental arch width data, the target tooth position includes the distal molar tooth positions on both sides of the maxilla and the distal molar tooth positions on both sides of the mandible, the distance training data includes the upper dental arch width parameters corresponding to the distal molar tooth positions on both sides of the maxilla, and the lower dental arch width parameters corresponding to the distal molar tooth positions on both sides of the mandible; the "calculating the average distance data and the training distance standard deviation of the distance training data" specifically includes: Calculate the upper average width value of all upper dental arch width parameters in all distance training data, and the lower average width value of all lower dental arch width parameters in all distance training data, to obtain the average distance data; The upper width standard deviation of all upper dental arch width parameters in all distance training data and the lower width standard deviation of all lower dental arch width parameters in all distance training data are calculated to obtain the training distance standard deviation.

14. The method for generating oral structure information according to claim 10, characterized in that: The distribution probability coefficient is 2.

15. The method for generating oral structure information according to claim 2, characterized in that: The "calculating and obtaining the characteristic distance value between the reference characteristic coordinates and the boundary characteristic coordinates" specifically includes: The distance value between the reference feature coordinate and the boundary feature coordinate in the extension direction of the tooth centerline is calculated to obtain the feature distance value.

16. The method for generating oral structure information according to claim 15, characterized in that: The first direction is the projection of the extension direction of the long axis of the tooth on the three-dimensional model of the oral cavity; the boundary feature coordinates and the reference feature coordinates are located in a preset model coordinate system, and the model coordinate system at least includes a first coordinate axis and a third coordinate axis, the third coordinate axis extends along the extension direction of the tooth midline, and the first coordinate axis extends along the extension direction of the central incisor width; The “calculating the distance between the reference feature coordinates and the boundary feature coordinates in the extension direction of the tooth centerline” specifically includes: The coordinate difference between the reference feature coordinates and the boundary feature coordinates on the third coordinate axis is calculated.

17. The method for generating oral structure information according to claim 2, characterized in that: The "comparing the characteristic distance value with the distance integrity criterion value characterizing the preset position condition, and judging whether the intraoral tissue area corresponding to the target tooth position in the oral three-dimensional model meets the preset integrity condition according to the comparison result" specifically includes: Determining whether the characteristic distance value is less than the distance integrity criterion value; If yes, then determining that the position in the oral three-dimensional model pointed to by the characteristic distance value is a feature missing position; If not, then continue to compare the next characteristic distance value with the distance integrity criterion value; According to the number of the feature missing positions, it is determined whether the intraoral tissue region corresponding to the target tooth position in the oral three-dimensional model satisfies the integrity condition.

18. The method for generating oral structure information according to claim 17, characterized in that: The "determining whether the intraoral tissue region corresponding to the target tooth position in the oral three-dimensional model satisfies the integrity condition according to the number of the feature missing positions" specifically includes: Determine the numerical relationship between the number of feature missing positions and the allowable error value; If the number of the feature missing positions is greater than or equal to the allowable error number value, it is determined that the intraoral tissue region corresponding to the target tooth position in the oral three-dimensional model does not meet the integrity condition; If the number of the feature missing positions is less than the allowable error number value, it is determined that the intraoral tissue region corresponding to the target tooth position in the oral three-dimensional model meets the integrity condition; The allowable error value is an integer greater than or equal to half of the target tooth position number.

19. The method for generating oral structure information according to claim 1, characterized in that: The "calculating the positional relationship between the maxillary model boundary and the mandibular model boundary according to the target tooth position features in the oral three-dimensional model data and the model boundary features corresponding to the target tooth position features" specifically includes: Determine at least the incisor tooth position on the oral three-dimensional model as the target tooth position, and calculate the reference feature coordinates of the target tooth position based on a preset feature recognition rule to characterize the target tooth position feature; According to the reference feature coordinates, determine the upper boundary extreme value coordinates and the lower boundary extreme value coordinates located in the oral three-dimensional model along the first direction; wherein the upper boundary extreme value coordinates are located in the upper jaw of the oral three-dimensional model, and the distance between the upper boundary extreme value coordinates and the occlusal plane has a maximum value, and the lower boundary extreme value coordinates are located in the lower jaw of the oral three-dimensional model, and the distance between the lower boundary extreme value coordinates and the occlusal plane has a maximum value; Calculating a characteristic height value between the upper boundary extreme value coordinates and the lower boundary extreme value coordinates to characterize the positional relationship between the maxillary model boundary and the mandibular model boundary; The "comparing the positional relationship with the preset positional condition corresponding to the positional relationship to determine whether the intraoral tissue region corresponding to the target tooth position in the oral three-dimensional model satisfies the preset integrity condition" specifically includes: The characteristic height value is compared with the height integrity criterion value characterizing the preset position condition, and based on the comparison result, it is determined whether the intraoral tissue area corresponding to the target tooth position in the oral three-dimensional model meets the preset integrity condition.

20. The method for generating oral structure information according to claim 19, characterized in that: The first direction is the projection of the extension direction of the long axis of the tooth on the jaw surface of the three-dimensional model of the oral cavity, the reference feature coordinates, the upper boundary extreme value coordinates and the lower boundary extreme value coordinates are located in a preset model coordinate system, the origin of the model coordinate system is located on the straight line where the incisor ridge is located, and at least includes a first coordinate axis and a third coordinate axis, the third coordinate axis extends along the extension direction of the tooth midline, and the first coordinate axis extends along the extension direction of the central incisor width; The “determining the upper boundary extreme value coordinates and the lower boundary extreme value coordinates located in the oral three-dimensional model along the first direction according to the reference feature coordinates” specifically includes: Calculating an upper boundary coordinate set corresponding to the maxillary target tooth position on the oral three-dimensional model, and a lower boundary coordinate set corresponding to the mandibular target tooth position on the oral three-dimensional model, so as to characterize the model boundary features; Traverse and determine the upper boundary coordinate with the maximum coordinate value corresponding to the third coordinate axis in the upper boundary coordinate set as the upper boundary extreme value coordinate, and traverse and determine the lower boundary coordinate with the minimum coordinate value corresponding to the third coordinate axis in the lower boundary coordinate set as the lower boundary extreme value coordinate; The "calculating and obtaining the characteristic height value between the upper boundary extreme value coordinates and the lower boundary extreme value coordinates" specifically includes: The coordinate difference between the upper boundary extreme value coordinate and the lower boundary extreme value coordinate on the third coordinate axis is calculated as the characteristic height value.

21. The method for generating oral structure information according to claim 19, characterized in that: Before the step of "comparing the characteristic height value with the height integrity criterion value characterizing the preset position condition, and judging whether the intraoral tissue area corresponding to the target tooth position in the oral three-dimensional model meets the preset integrity condition according to the comparison result", the step includes: Acquire at least two sets of three-dimensional training model data, and at least two sets of height training data corresponding to the three-dimensional training model data; wherein the height training data includes the distance between the upper tissue extreme value coordinates on the three-dimensional training model corresponding to the intraoral tissue area on one side of the first direction and the lower tissue extreme value coordinates on the three-dimensional training model corresponding to the intraoral tissue area on the other side of the first direction; Calculate the average height data and the training height standard deviation of the height training data, and calculate the height integrity criterion value based on the difference between the product of the average height data and the training height standard deviation and a preset distribution probability coefficient; wherein the distribution probability coefficient is 3.

22. The method for generating oral structure information according to claim 1, characterized in that: The “extracting oral feature data corresponding to the intraoral tissue area from the intraoral image data” specifically includes: Calling the corresponding first neural network model and second neural network model according to the intraoral tissue area; Inputting the intraoral image data into the first neural network model to extract the region of interest, thereby obtaining feature region image data; The characteristic region image data is input into the second neural network model for feature recognition to obtain the oral characteristic data corresponding to the intraoral tissue region.

23. The method for generating oral structure information according to claim 22, characterized in that: The first neural network model and the second neural network model are convolutional neural network models; the intraoral tissue area includes the vestibular sulcus; before the step of “calling the corresponding first neural network model and second neural network model according to the intraoral tissue area”, the step includes: Acquire at least two sets of training image data and region of interest marks corresponding to the training image data; wherein the extension range of the region of interest marks in the corresponding training image at least covers the upper lip, lower lip and dentition in the training image; Calling a preset convolutional neural network model, performing iterative training with the training image data and the region of interest mark as model input, and obtaining first training parameters and a corresponding first neural network model; Acquire at least two sets of training image data, and tooth position marks and vestibular groove bottom marks corresponding to the training image data; The preset convolutional neural network model is called, and iterative training is performed using the training image data, the tooth position mark and the vestibule groove bottom mark as model input to obtain second training parameters and a corresponding second neural network model.

24. The method for generating oral structure information according to claim 1, characterized in that: The at least one same direction includes a tooth crown width direction; the size mapping factor is a quotient of the reference physical size data and the reference pixel size data.

25. The method for generating oral structure information according to claim 24, characterized in that: The "calculating the dimension data of the tooth crown at the relative reference tooth position in at least one same direction in the intraoral image and the oral three-dimensional model, and obtaining reference pixel dimension data and reference physical dimension data respectively" specifically includes: Determine the first reference feature point and the second reference feature point of the relative reference tooth position on the intraoral image, calculate and obtain the reference pixel size data based on the number of pixels between the first reference feature point and the second reference feature point; wherein the first reference feature point and the second reference feature point are located on both sides of the long axis of the tooth of the relative reference tooth position, and the distance between the first reference feature point and the corresponding incisal end of the tooth crown of the relative reference tooth position is equal to the distance between the second reference feature point and the corresponding incisal end of the tooth crown of the relative reference tooth position; The third reference feature point and the fourth reference feature point of the relative reference tooth position are determined on the three-dimensional model of the oral cavity, and the reference physical dimension data is obtained by calculating and based on the Euclidean distance between the third reference feature point and the fourth reference feature point; wherein the third reference feature point and the fourth reference feature point are located on both sides of the long axis of the tooth of the relative reference tooth position, and the distance between the third reference feature point and the corresponding relative reference tooth position crown incisal end is equal to the distance between the fourth reference feature point and the corresponding relative reference tooth position crown incisal end.

26. The method for generating oral structure information according to claim 25, characterized in that: The "determining at least one target tooth position corresponding to each other in the intraoral image and the oral three-dimensional model as a relative reference tooth position" specifically includes: Determine whether the maxillary left central incisor tooth position and the maxillary right central incisor tooth position in the intraoral image and the oral three-dimensional model both contain complete crowns; If yes, then determining the maxillary left central incisor tooth position and the maxillary right central incisor tooth position corresponding to each other in the intraoral image and the oral three-dimensional model as relative reference tooth positions; The "determining the first reference feature point and the second reference feature point of the relative reference tooth position on the intraoral image, and calculating and obtaining the reference pixel size data according to the number of pixels between the first reference feature point and the second reference feature point" specifically includes: Determine the first reference feature point at the distal boundary of the crown of the maxillary left central incisor tooth position in the intraoral image, and determine the second reference feature point at the distal boundary of the crown of the maxillary right central incisor tooth position in the intraoral image; The number of pixels between the first reference feature point and the second reference feature point is calculated, and half of the number of pixels is used as the reference pixel size data.

27. The method for generating oral structure information according to claim 26, characterized in that: The distance from the first reference feature point to the gingival end of the corresponding incisor tooth position is three times the distance from the first reference feature point to the incisal end of the crown of the corresponding incisor tooth position.

28. The method for generating oral structure information according to claim 25, characterized in that: The "determining at least one target tooth position corresponding to each other in the intraoral image and the oral three-dimensional model as a relative reference tooth position" specifically includes: Determine whether the maxillary left central incisor tooth position and the maxillary right central incisor tooth position in the intraoral image and the oral three-dimensional model both contain complete crowns; If only the maxillary left central incisor tooth position in the intraoral image and the oral three-dimensional model contains a complete tooth crown, then determining the maxillary left central incisor tooth position corresponding to each other in the intraoral image and the oral three-dimensional model as the relative reference tooth position; The "determining the first reference feature point and the second reference feature point of the relative reference tooth position on the intraoral image, and calculating and obtaining the reference pixel size data according to the number of pixels between the first reference feature point and the second reference feature point" specifically includes: Determine the first reference feature point at the distal boundary of the crown of the maxillary left central incisor tooth position in the intraoral image, and determine the second reference feature point at the mesial boundary of the crown of the maxillary left central incisor tooth position in the intraoral image; The number of pixels between the first reference feature point and the second reference feature point is calculated, and the number of pixels is used as the reference pixel size data.

29. The method for generating oral structure information according to claim 1, characterized in that: The intraoral tissue area includes the vestibule groove, the oral cavity feature data includes vestibule groove height data, and the intraoral tissue feature points include upper groove bottom feature points and lower groove bottom feature points.

30. The method for generating oral structure information according to claim 29, characterized in that: The "fitting the tissue space feature points corresponding to the intraoral tissue feature points in accordance with the pixel size relationship" specifically includes: Calculate the distance between the upper groove bottom feature point and the midpoint of the gingival margin of the corresponding tooth position to obtain the upper groove bottom spacing parameter, and calculate the corresponding upper groove bottom mapping parameter according to the upper groove bottom spacing parameter and the size mapping factor representing the pixel size relationship; According to the reference characteristic coordinates of the target tooth position corresponding to the upper groove bottom characteristic point and the upper groove bottom mapping parameters, an upper spatial characteristic point characterizing the upper groove bottom position is fitted; Calculate the distance between the lower groove bottom feature point and the midpoint of the gingival margin of the corresponding tooth position to obtain the lower groove bottom spacing parameter, and calculate the corresponding lower groove bottom mapping parameter according to the lower groove bottom spacing parameter and the size mapping factor; According to the reference characteristic coordinates of the target tooth position corresponding to the lower groove bottom characteristic point and the lower groove bottom mapping parameters, a lower spatial characteristic point characterizing the position of the lower groove bottom is obtained by fitting.

31. The method for generating oral structure information according to claim 1, characterized in that: The intraoral tissue area includes the vestibule groove and / or the tooth root ridge, the oral cavity feature data includes dental arch width data, and the intraoral tissue feature points include left groove bottom feature points and right groove bottom feature points.

32. The method for generating oral structure information according to claim 1, characterized in that: The “determining at least two intraoral tissue feature points in the intraoral image data according to the oral cavity feature data” specifically includes: In the intraoral image, determining all intraoral tissue feature points according to the oral cavity feature data; The “part representing the oral cavity feature data on the oral cavity three-dimensional model data” specifically includes: A tissue space distribution curve or a tissue space distribution surface is fitted according to the tissue space feature points, and this is used as a part of the oral cavity three-dimensional model data that represents the oral cavity feature data.

33. A method for generating oral structure information, characterized in that: include: Acquire oral three-dimensional model data and corresponding intraoral image data; Extracting oral feature data corresponding to a target oral tissue region in the oral three-dimensional model from the intraoral image data, establishing a pixel size relationship between the intraoral image data and the oral three-dimensional model data, and reconstructing a portion representing the oral feature data on the oral three-dimensional model data to obtain the oral structure information; The “reconstructing the part representing the oral characteristic data on the oral three-dimensional model data to obtain the oral structure information” specifically includes: In the intraoral image data, determining at least two intraoral tissue feature points according to the oral cavity feature data; According to the intraoral tissue feature points, at the feature missing position in the oral three-dimensional model, tissue space feature points corresponding to the intraoral tissue feature points are fitted according to the pixel size relationship, and a part representing the oral feature data is formed on the oral three-dimensional model data; The “establishing a pixel size relationship between the intraoral image data and the oral three-dimensional model data” specifically includes: Determine at least one target tooth position corresponding to each other in the intraoral image and the oral three-dimensional model as a relative reference tooth position, and calculate the size data of the crown of the relative reference tooth position in at least one same direction in the intraoral image and the oral three-dimensional model, and obtain reference pixel size data and reference physical size data respectively; A size mapping factor is obtained by fitting according to the reference pixel size data and the reference physical size data to characterize the pixel size relationship.

34. An oral structure information generation system, comprising a processor, a memory and a communication bus, characterized in that: The processor and the memory communicate with each other via the communication bus; The memory is used to store the application program; The processor is used to implement the steps of the oral structure information generation method described in any one of claims 1-33 when executing the application stored in the memory.

35. A storage medium having an application stored thereon, characterized in that: When the application is executed, the steps of the oral structure information generation method described in any one of claims 1-33 are implemented.

36. An oral appliance, characterized in that: The oral appliance is constructed according to oral structure information. Generated according to the oral structure information generation method according to any one of claims 1-33.

37. The oral appliance of claim 36, wherein: The oral appliance is used for training the orofacial muscle function and / or for treating mouth breathing.

Images