Method and system for eliminating alveolar lines in gingival segmentation model extracted by oral scanning model

Abstract

The invention discloses a method and system for eliminating an alveolar line in a gingival segmentation model extracted by a mouth scanning model, and the method comprises the steps: constructing a three-dimensional model of gingival segmentation based on a surface curvature analysis method, and carrying out the projection and point cloud processing, thereby obtaining a homogenized pseudo three-dimensional model; performing sag compensation and gap filling processing to obtain a preprocessed pseudo three-dimensional model; sequentially carrying out region filling on the preprocessed pseudo three-dimensional model through a Flood Fill algorithm to obtain a plurality of region blocks with identifiers; and carrying out tooth groove line feature judgment, and carrying out combined filling and tooth groove line elimination processing to obtain a curvature feature three-dimensional model after tooth groove line elimination. According to the method, the efficiency and accuracy of tooth three-dimensional model construction can be improved, the tooth groove features are effectively eliminated, and the accuracy of extracting the region of each tooth is improved. The method and the system for eliminating the alveolar lines in the gingival segmentation model extracted by the oral scanning model can be widely applied to the technical field of image recognition and segmentation.

Description

Technical Field

[0001] The present invention relates to the technical field of image recognition and segmentation, and in particular to a method and system for eliminating tooth groove lines in a gingival segmentation model extracted from an oral scan model. Background Art

[0002] Tooth segmentation is one of the key technologies in the field of computer-assisted orthodontics and dental implants. In oral CAD / CAM systems, in order to accurately measure tooth parameters, simulate tooth movement, and rearrange teeth, it is first necessary to separate the single tooth model from the three-dimensional dental model. A common method is to filter the three-dimensional model of the feature area based on curvature information, and apply morphological algorithms to extract the dentition feature lines. Subsequently, these feature lines are used to segment the tooth model, thereby achieving independent analysis and processing of each tooth. However, when using the method of filtering feature areas based on curvature information, the uneven occlusal surface texture (tooth groove line) on the occlusal surface of the tooth is often mistakenly selected as the feature area for cutting teeth and gums due to its large curvature variation, which will greatly interfere with the separation of the single tooth model from the three-dimensional dental model, resulting in inaccurate or even erroneous segmentation results. Summary of the Invention

[0003] In order to solve the above technical problems, the purpose of the present invention is to provide a method and system for eliminating groove lines in the gingival segmentation model extracted from an oral scanning model, which can improve the efficiency and accuracy of constructing a three-dimensional tooth model, effectively eliminate the groove features, and improve the accuracy of the extraction of the area of each tooth.

[0004] The first technical solution adopted by the present invention is: a method for eliminating tooth groove lines in a gingival segmentation model extracted from an oral scan model, comprising the following steps:

[0005] Based on the surface curvature analysis method, a 3D model of tooth and gum segmentation is constructed and then projected and converted into a point cloud to obtain a homogenized pseudo 3D model.

[0006] The homogenized pseudo 3D model is sequentially subjected to concave compensation and gap filling processing to obtain a pre-processed pseudo 3D model;

[0007] The pre-processed pseudo 3D model is sequentially filled with regions using the FloodFill algorithm to obtain several region blocks with labels.

[0008] The tooth groove line features of several marked area blocks are judged, and the tooth groove line is merged, filled and eliminated. After inverse mapping, the curvature feature three-dimensional model after eliminating the tooth groove line is obtained.

[0009] Furthermore, the step of constructing a three-dimensional model of tooth and gum segmentation based on the surface curvature analysis method and performing projection and point cloud processing to obtain a homogenized pseudo three-dimensional model specifically includes:

[0010] Based on the surface curvature analysis method, the feature area recognition and processing of the target oral and maxillary 3D scanning model are performed to construct a 3D model of tooth and gum segmentation;

[0011] Based on the occlusal surface of the target oral dental and maxillary 3D scanning model as the top view, the 3D coordinates in the tooth and gum segmentation 3D model are projected into 2D coordinates to construct a pseudo 3D model;

[0012] The target point cloud density of the pseudo 3D model is obtained, the plane triangle coverage area of the pseudo 3D model is determined and point cloud processing is performed to obtain a homogenized pseudo 3D model.

[0013] Furthermore, the step of performing feature area recognition processing on the target oral dental and jaw 3D scan model based on the surface curvature analysis method to construct a 3D tooth and gum segmentation model specifically includes:

[0014] Scan the target oral cavity with a scanner to build a three-dimensional model of the jaw;

[0015] Obtain the average surface curvature of the three-dimensional dental model and set the curvature threshold;

[0016] The parts of the surface with an average curvature greater than a curvature threshold in the three-dimensional model of the jaw are marked as feature areas and extracted to construct a three-dimensional model of tooth and gum segmentation.

[0017] Furthermore, the step of sequentially performing concave compensation and gap filling processing on the homogenized pseudo 3D model to obtain a pre-processed pseudo 3D model specifically includes:

[0018] Perform data analysis on the point cloud data in the homogenized pseudo 3D model, set the number of detection distance points in the horizontal and vertical directions, obtain and mark the sparse part of the point cloud data, and determine the point cloud data to be compensated;

[0019] Through the expansion algorithm, a new point cloud is inserted into the point cloud data to be compensated according to the target point cloud density to obtain a compensated pseudo 3D model;

[0020] Scanning the compensated pseudo three-dimensional model to obtain gaps in the compensated pseudo three-dimensional model and extracting all endpoints at both ends of all gaps;

[0021] Establish key point pairs at both ends of all endpoints through spatial position relationship matching algorithm;

[0022] The key point pairs are continuously patched using a patching algorithm to obtain a pre-processed pseudo three-dimensional model.

[0023] Furthermore, the step of sequentially filling regions of the pre-processed pseudo 3D model using a Flood Fill algorithm to obtain a plurality of region blocks with identifiers specifically includes:

[0024] Obtaining the closed area of the preprocessed pseudo three-dimensional model;

[0025] Based on the Flood Fill algorithm, the starting point and filling threshold are determined, and the multiple closed areas of the pre-processed pseudo 3D model are filled to obtain several area blocks;

[0026] Marking a number of area blocks to obtain a number of area blocks with identifiers.

[0027] Furthermore, the step of determining the tooth groove line features of the marked area blocks, performing merging, filling, and eliminating tooth groove line processing, and obtaining a curvature feature three-dimensional model after eliminating the tooth groove line through inverse mapping specifically includes:

[0028] Performing tooth groove line feature judgment on several marked area blocks and merging them to obtain merged area blocks;

[0029] Filling the gaps in the merged area blocks by a gap filling method to obtain area blocks after eliminating the tooth groove lines;

[0030] The area block after the groove line is eliminated is inversely mapped to the tooth and gum segmentation three-dimensional model to obtain the curvature feature three-dimensional model after the groove line is eliminated.

[0031] Furthermore, the step of determining the tooth groove line features of the marked area blocks and merging them to obtain the merged area blocks specifically includes:

[0032] Obtaining the geometric center, maximum vertical coordinate value and minimum vertical coordinate value of a plurality of area blocks with identification;

[0033] Determine the coordinate difference of each area block according to the maximum vertical coordinate value and the minimum vertical coordinate value;

[0034] If the maximum vertical coordinate value and the minimum vertical coordinate value of the two area blocks overlap, obtain the length of the overlap and compare the coordinate difference between the two area blocks, and select the area block with the smallest coordinate difference as the reference area;

[0035] Set the threshold feature of the inter-tooth overlap of the molar part of the teeth;

[0036] The ratio data between the overlapping length and the coordinate difference of the reference area is calculated and judged with the threshold feature of the inter-tooth overlap of the molar part of the tooth. If the ratio data is greater than the threshold feature of the inter-tooth overlap of the molar part of the tooth, the two area blocks are merged to obtain a merged area block.

[0037] Furthermore, the step of inversely mapping the area block after the groove lines are eliminated to the tooth and gum segmentation three-dimensional model to obtain the curvature feature three-dimensional model after the groove lines are eliminated specifically includes:

[0038] Deleting the tooth region identifier in the region block after the tooth groove line is eliminated to obtain a pre-processed region block;

[0039] Based on the three-dimensional model of tooth and gum segmentation corresponding to the preprocessed area block, each triangular face in the three-dimensional model is scanned sequentially, the coordinate points of the triangular face are projected, and the triangular faces whose coordinate points fall in the area block are retained to obtain a three-dimensional model with curvature characteristics after eliminating the groove lines.

[0040] The second technical solution adopted by the present invention is: a system for eliminating tooth groove lines in a gingival segmentation model extracted from an oral scan model, comprising:

[0041] The first module is used to construct a 3D model of tooth and gum segmentation based on the surface curvature analysis method and perform projection and point cloud processing to obtain a homogenized pseudo 3D model;

[0042] The second module is used to perform concave compensation and gap filling processing on the homogenized pseudo 3D model in sequence to obtain a pre-processed pseudo 3D model;

[0043] The third module is used to sequentially fill regions of the pre-processed pseudo 3D model using a Flood Fill algorithm to obtain a number of region blocks with identifiers;

[0044] The fourth module is used to judge the groove line features of several marked area blocks, merge and fill them, and eliminate the groove lines. After inverse mapping, a three-dimensional model of the curvature features after eliminating the groove lines is obtained.

[0045] The beneficial effects of the method and system of the present invention are as follows: the present invention constructs a three-dimensional model of tooth and gum segmentation through a surface curvature analysis method and performs projection and point cloud processing to obtain a homogenized pseudo three-dimensional model; by homogenizing the point cloud density, the point cloud processing efficiency is improved, ensuring that subsequent analysis can be performed on the basis of point cloud data with consistent density; further, the homogenized pseudo three-dimensional model is sequentially subjected to concave compensation and gap filling processing, which can effectively reduce the risk of information loss caused by point cloud projection, ensuring that the algorithm complexity or running time will not be increased due to sparse or concave point clouds during subsequent analysis; and by judging and identifying gaps in the pseudo three-dimensional model, point pairs are formed at both ends of the gap, and repairs are performed between the point pairs, thereby restoring the integrity of the model, and through Flood The Fill algorithm fills the pre-processed pseudo three-dimensional model in sequence, distinguishes different areas in the model, and labels each area. Finally, it judges the groove line features of several marked area blocks, merges and fills them, and eliminates the groove lines. It uses the high longitudinal overlap features of the blocks to integrate multiple independent areas into an overall area, thereby improving the efficiency and accuracy of the three-dimensional tooth model processing, effectively eliminating the interference of groove features on model calculation and analysis, and ensuring that the area of each tooth can be extracted more accurately. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 This is a flowchart of the steps of a method for eliminating the tooth groove line in a gingival segmentation model extracted from an oral scanning model according to the present invention;

[0047] Figure 2 This is a structural block diagram of a system for eliminating tooth groove lines in a gingival segmentation model extracted from an oral scanning model according to the present invention;

[0048] Figure 3 is a schematic diagram of a three-dimensional model of a dental jaw provided in a specific embodiment of the present invention;

[0049] Figure 4 is a schematic diagram of calculating the average curvature of the surface of a three-dimensional dental model provided by a specific embodiment of the present invention;

[0050] Figure 5 is a schematic diagram of a three-dimensional model of teeth and gums segmentation provided by a specific embodiment of the present invention;

[0051] Figure 6 is a schematic diagram of a homogenized pseudo three-dimensional model provided by a specific embodiment of the present invention;

[0052] Figure 7 is a schematic diagram of a compensated pseudo three-dimensional model provided by a specific embodiment of the present invention;

[0053] Figure 8Schematic diagram of determining the spatial position relationship of key point pairs at both ends of a gap provided by a specific embodiment of the present invention;

[0054] Figure 9 is a schematic diagram of a preprocessed pseudo three-dimensional model provided by a specific embodiment of the present invention;

[0055] Figure 10 This is a schematic diagram of merging multiple regions corresponding to teeth with groove features into an overall region, provided by a specific embodiment of the present invention;

[0056] Figure 11 is a schematic diagram of gap filling provided by a specific embodiment of the present invention using the marking of the tooth area for filling;

[0057] Figure 12 It is a schematic diagram of a three-dimensional model of curvature characteristics after eliminating the groove line provided by a specific embodiment of the present invention. DETAILED DESCRIPTION

[0058] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. The step numbers in the following embodiments are provided for ease of description only and do not limit the order of the steps. The order of execution of the steps in the embodiments can be adaptively adjusted according to the understanding of those skilled in the art.

[0059] First of all, it should be noted that the importance of eliminating the sulcus feature area lies in the fact that accurate tooth and gum segmentation is the basis for subsequent analysis and processing. For example, in tooth morphological studies, orthodontic treatment planning, dental implants, and dental health assessments, accurate segmentation results can provide more reliable data support. If the sulcus feature area is not effectively eliminated, it may lead to misdiagnosis or wrong treatment plans, thereby affecting the patient's treatment effect and oral health. Therefore, ensuring the accuracy of the segmentation process and eliminating those irrelevant or interfering feature areas is one of the important steps to obtain high-quality tooth models and carry out subsequent analysis and treatment. Its accuracy is directly related to the effect and efficiency of subsequent diagnosis and treatment.

[0060] Based on this, the embodiment of the present invention extracts a three-dimensional model of the gums and teeth segmented from a region with a large curvature, which contains a groove feature area, based on the surface curvature analysis of the three-dimensional jaw model. First, the segmented three-dimensional model is projected from the positive side of the occlusal surface. Then, the concave part of the projected three-dimensional model is filled by the expansion method. Next, the sliding window detection method is used to perform connectivity repair on the gap part. Then, multiple rounds of Flood Fill algorithm are used to calculate the area of multiple teeth. The area of the same tooth with a groove line is determined and merged through the regional directional overlap determination strategy. Finally, the gap of the merged tooth area is filled to obtain a complete area of the tooth with a groove line. This solves the problem that the curvature feature area formed by the groove when the three-dimensional jaw model uses curvature features to segment teeth interferes with tooth extraction, and significantly improves the accuracy of tooth extraction.

[0061] Reference Figure 1 The present invention provides a method for eliminating tooth groove lines in a gingival segmentation model extracted from an oral scan model, the method comprising the following steps:

[0062] S100, constructing a three-dimensional model of tooth and gum segmentation based on a surface curvature analysis method and performing projection and point cloud processing to obtain a homogenized pseudo three-dimensional model;

[0063] S110, performing feature region recognition processing on the target oral dental and jaw 3D scan model based on a surface curvature analysis method, and constructing a 3D tooth and gum segmentation model;

[0064] Among them, the target oral cavity is scanned and processed by a scanner to construct a three-dimensional model of the jaw; the surface average curvature of the three-dimensional model of the jaw is obtained, and the curvature threshold is set; the part of the three-dimensional model of the jaw where the surface average curvature is greater than the curvature threshold is marked as a feature area and extracted and processed to construct a three-dimensional model of tooth and gum segmentation.

[0065] In this embodiment, a scanner is used to scan the oral cavity to obtain a three-dimensional model of the jaws, such as Figure 3 As shown, further, the average curvature of the surface of the three-dimensional model of the tooth jaw is calculated. The calculation method is to take the maximum and minimum curvature values k1 and k2 of each point and calculate the average of the two as the average curvature of the point. The average curvature of each point is marked with red, blue and green from high to low, as shown in Figure 1. Figure 4 As shown, Figure 4 The red part is where the calculated curvature is larger, and then the points with larger curvature values on the surface of the 3D model are marked in red. The red area is the feature area, and finally the feature area is extracted to form a 3D model of the teeth and gums segmentation, such as Figure 5As shown, the red area is the groove line of the two molars. When this model is used to segment and cut the original model, large defects will appear on the surface of the three-dimensional model of the two molars. At this time, most of the new three-dimensional model is located on the tooth-gingival boundary line and the adjacent tooth boundary line. However, there are also feature areas in the tooth occlusal surface area, which will interfere with the use of curvature feature areas to segment the three-dimensional model of the tooth part.

[0066] By using a scanner to acquire a 3D model of the jaw, calculating its surface curvature, and identifying characteristic areas with large curvatures, a new 3D model is ultimately formed. This process not only provides important feature information for subsequent analysis but also lays the foundation for precise processing in subsequent steps.

[0067] S120, based on the occlusal surface of the target oral dental and maxillary 3D scan model as a top view, projecting the 3D coordinates in the tooth and gum segmentation 3D model into 2D coordinates to construct a pseudo 3D model;

[0068] S130 , obtaining a target point cloud density of the pseudo 3D model, determining a plane triangle coverage area of the pseudo 3D model, and performing point cloud processing to obtain a homogenized pseudo 3D model.

[0069] In this embodiment, when generating a pseudo three-dimensional model, the position and direction of the projection plane and the type of projection algorithm are first determined. According to the set projection plane, the three-dimensional coordinates are converted into two-dimensional coordinates to form a pseudo three-dimensional model. This process projects the triangle ΔABC in the three-dimensional space to the triangle ΔA'B'C' on the two-dimensional plane. A mathematical model is used for projection calculation to ensure the accuracy of the projection, and then the target point cloud density D is determined. The coverage area of the triangle ΔA'B'C' is processed into a point cloud according to the density D. Finally, a visualization tool is used to check the generated pseudo three-dimensional model and the uniformized point cloud effect to ensure the integrity and details of the model. After the construction of the embodiment of the present invention is completed, it contains 263,653 points, such as Figure 6 shown.

[0070] Projecting the 3D model into a pseudo 3D model improves the point cloud processing efficiency by homogenizing the point cloud density, ensuring that subsequent analysis can be performed on point cloud data with consistent density, thereby optimizing the overall processing flow.

[0071] S200, performing concave compensation and gap filling processing on the homogenized pseudo 3D model in sequence to obtain a pre-processed pseudo 3D model;

[0072] S210, performing data analysis on the point cloud data in the homogenized pseudo 3D model, setting the number of detection distance points in the horizontal and vertical directions, obtaining and marking sparse parts of the point cloud data, and determining the point cloud data to be compensated;

[0073] In this embodiment, the point cloud data is analyzed, the number of detection distance points N1 in the horizontal and vertical directions is set, the sparse or concave parts of the point cloud are identified, and the point cloud parts that need to be compensated {p1, p2, ..., p n}.

[0074] S220, using an expansion algorithm, inserting a new point cloud into the point cloud data to be compensated according to the target point cloud density to obtain a compensated pseudo 3D model;

[0075] In this embodiment, for the part of the point cloud that needs to be compensated, the distribution of the surrounding point clouds is taken into consideration, and an expansion algorithm is used to insert new point clouds according to a given point cloud density D to compensate for the sparse or concave part of the point cloud. After compensation, the point cloud increases to 853487 points. Figure 7 This effectively reduces the risk of information loss due to point cloud projection, ensuring that subsequent analysis will not increase algorithm complexity or extend runtime due to sparse or concave point clouds.

[0076] S230, scanning the compensated pseudo three-dimensional model to obtain gaps in the compensated pseudo three-dimensional model, and extracting all endpoints at both ends of all gaps;

[0077] In this embodiment, the pseudo 3D model is fully scanned and analyzed, and a detection algorithm is used to identify the gaps in the model and accurately locate the key points at both ends of the gaps.<v1,v2> All the key points at both ends of the gap form point pairs. By judging and identifying the gaps in the pseudo three-dimensional model, point pairs at both ends of the gap are formed, and repairs are performed between the point pairs to restore the integrity of the model. This is the key to subsequent region extraction.

[0078] S240, establishing key point pairs at both ends of all endpoints through a spatial position relationship matching algorithm;

[0079] In this embodiment, a matching algorithm is used to compare the point pairs at both ends of the gap.<v1,v2> , to ensure that the position relationship of the selected point pairs in space is reasonable, that is, the point pairs determined<v1,v2> Located at both ends of the gap in the 3D model of tooth and gum segmentation, such as Figure 8 The positional relationship of the endpoints demarcated by the two red boxes in space is reasonable.

[0080] S250: Perform continuous patching processing on the key point pairs using a patching algorithm to obtain a pre-processed pseudo three-dimensional model.

[0081] In this embodiment, a patching algorithm is implemented based on the identified and matched point pairs at both ends of the gap to fill the gap and ensure a smooth transition between the newly generated point cloud and the surrounding structure. The patching process needs to consider the overall consistency and continuity of the point cloud to ensure that there is no obvious abruptness between the geometric features of the patched area and the original model. The point cloud after the gap is filled contains 1170283 points, such as Figure 9 shown.

[0082] S300, using the FloodFill algorithm, sequentially filling regions of the pre-processed pseudo 3D model to obtain a number of region blocks with identifiers;

[0083] Specifically, a closed area of the preprocessed pseudo 3D model is obtained; based on the Flood Fill algorithm, a starting point and a filling threshold are determined, and multiple closed areas of the preprocessed pseudo 3D model are filled to obtain a number of area blocks; and the number of area blocks are marked to obtain a number of area blocks with identification.

[0084] In this embodiment, based on the pre-processed pseudo 3D model, it is possible to ensure that the Flood Fill algorithm can accurately identify each area in the model. The starting point and filling threshold of the Flood Fill algorithm are determined so that different areas can be effectively distinguished when performing segmentation, ensuring that each area can be correctly identified and marked. Since there are multiple closed areas (A1, A2, ..., A m ), so the Flood Fill algorithm needs to be called multiple times to fill different areas. The segmented blocks are then identified, ensuring each area has a unique identifier for subsequent analysis and processing. This prevents confusion between blocks and allows for the distinction and labeling of different regions within the model. This provides a foundation for subsequent analysis to eliminate the influence of characteristic sulcus areas and helps extract clear, independent regions.

[0085] S400: determining the tooth groove line features of the marked area blocks, performing merging, filling and tooth groove line elimination processing, and obtaining a curvature feature three-dimensional model after eliminating the tooth groove lines through inverse mapping.

[0086] S410, determining the tooth groove line features of the plurality of marked area blocks, and merging them to obtain a merged area block;

[0087] Specifically, the geometric center, maximum vertical coordinate value when looking down and minimum vertical coordinate value when looking down of several area blocks with labels are obtained; the coordinate difference of each area block is determined according to the maximum vertical coordinate value when looking down and the minimum vertical coordinate value when looking down; if there is an overlap between the maximum vertical coordinate value and the minimum vertical coordinate value when looking down between two area blocks, the length of the overlap is obtained and the coordinate difference between the two area blocks is compared, and the area block corresponding to the smallest coordinate difference is selected as the reference area; a threshold feature of the inter-tooth overlap of the molar part of the teeth is set; the ratio data between the length of the overlap and the coordinate difference of the reference area is calculated, and the data is judged with the threshold feature of the inter-tooth overlap of the molar part of the teeth; if the ratio data is greater than the threshold feature of the inter-tooth overlap of the molar part of the teeth, the two area blocks are merged to obtain a merged area block.

[0088] In this embodiment, the geometric center of each region (C1, C2, ..., C m ), and the maximum and minimum vertical coordinate values. If two areas A i and A j The maximum and minimum vertical coordinate intervals of the two areas overlap, and the area with the smaller difference between the geometric center and the maximum and minimum vertical coordinates of the area is used as the reference area A. k , and the above difference is diff, and then calculate the two areas A i and A j The overlap length L occurs in the vertical coordinate interval when viewed from above. Finally, according to the inter-tooth overlap characteristics of the molar part of the teeth, the threshold V is set to 0.35. If L / diff exceeds the given threshold V, it is determined that the two regions A i and A j Belong to the same tooth and merge them. Merge multiple areas corresponding to teeth with groove characteristics into one overall area, such as Figure 10 As shown in the figure, the high longitudinal overlap of the blocks is used to integrate multiple independent regions into a single overall region. This merging process ensures that regions with sulcus features can be more effectively identified and processed in subsequent steps.

[0089] S420, filling the gaps in the merged area block using a gap filling method to obtain an area block after eliminating the groove lines;

[0090] In this embodiment, if Figure 11 As shown in the figure, the gap filling is carried out in both horizontal and vertical directions. The filling judgment is based on the model area ( Figure 11(red in the middle) The two ends of the image are marked tooth regions, and they are marked by the same tooth. Gap filling uses the markings of the tooth regions to fill the gaps, thereby identifying these areas as the tooth region. The merged regions are further processed to fill the gaps between the multiple regions to eliminate the influence of the tooth groove lines. This step lays the foundation for the final feature extraction.

[0091] S430 , inversely mapping the area block after the groove lines are eliminated to the tooth-gingival segmentation three-dimensional model to obtain a curvature feature three-dimensional model after the groove lines are eliminated.

[0092] Specifically, the tooth area identifier in the area block after the groove line is eliminated is deleted to obtain the preprocessed area block; based on the tooth and gum segmentation three-dimensional model corresponding to the preprocessed area block, each triangular face in the three-dimensional model is scanned sequentially, the coordinate points of the triangular face are projected, and the triangular faces whose coordinate points fall in the area block are retained to obtain the curvature feature three-dimensional model after the groove line is eliminated.

[0093] In this embodiment, before performing the inverse mapping calculation, all the identification features of different tooth regions need to be eliminated. The inverse mapping calculation takes the segmented three-dimensional model as the object and sequentially scans each triangular face in the three-dimensional model. For the current triangular face, it is determined whether the projection coordinates of its three vertices are within the obtained model area ( Figure 11 As long as there is a vertex in the area, the triangle is retained. All the retained triangles are reconstructed into a 3D model, that is, the segmented 3D model after eliminating the groove line is obtained, as shown in Figure 12 shown.

[0094] Reference Figure 2 , a system for eliminating tooth groove lines in a gingival segmentation model extracted from an oral scan model, comprising:

[0095] The first module 201 is used to construct a three-dimensional model of tooth and gum segmentation based on a surface curvature analysis method and perform projection and point cloud processing to obtain a homogenized pseudo three-dimensional model;

[0096] The second module 202 is used to perform concave compensation and gap filling processing on the homogenized pseudo 3D model in sequence to obtain a pre-processed pseudo 3D model;

[0097] The third module 203 is used to sequentially fill regions of the pre-processed pseudo 3D model using a Flood Fill algorithm to obtain a plurality of region blocks with identifiers;

[0098] The fourth module 204 is used to determine the dental groove line features of the marked area blocks, merge and fill them, and eliminate the dental groove lines, and obtain a three-dimensional model of the curvature features after eliminating the dental groove lines through inverse mapping.

[0099] The contents of the above method embodiments are all applicable to the present system embodiments. The functions specifically implemented by the present system embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0100] The above is a specific description of the preferred implementation of the present invention, but the invention is not limited to the embodiments. Those skilled in the art can make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of this application.

Claims

1. A method for eliminating tooth groove lines in a gingival segmentation model extracted from an oral scan model, characterized in that: The following steps are involved: Based on the surface curvature analysis method, a 3D model of tooth and gum segmentation is constructed and then projected and converted into a point cloud to obtain a homogenized pseudo 3D model. The homogenized pseudo 3D model is sequentially subjected to concave compensation and gap filling processing to obtain a pre-processed pseudo 3D model; The pre-processed pseudo 3D model is sequentially filled with regions using the Flood Fill algorithm to obtain several region blocks with labels. The tooth groove line features of several marked area blocks are judged, and the tooth groove line is merged, filled and eliminated. After inverse mapping, the curvature feature three-dimensional model after eliminating the tooth groove line is obtained.

2. The method for eliminating sulcus lines in a gingival segmentation model extracted from an oral scan model according to claim 1, characterized in that: The step of constructing a tooth and gum segmentation three-dimensional model based on the surface curvature analysis method and performing projection and point cloud processing to obtain a homogenized pseudo three-dimensional model specifically includes: Based on the surface curvature analysis method, the feature area recognition and processing of the target oral and maxillary 3D scanning model are performed to construct a 3D model of tooth and gum segmentation; Based on the occlusal surface of the target oral dental and maxillary 3D scanning model as the top view, the 3D coordinates in the tooth and gum segmentation 3D model are projected into 2D coordinates to construct a pseudo 3D model; The target point cloud density of the pseudo 3D model is obtained, the plane triangle coverage area of the pseudo 3D model is determined and point cloud processing is performed to obtain a homogenized pseudo 3D model.

3. The method for eliminating groove lines in a gingival segmentation model extracted from an oral scan model according to claim 2, characterized in that: The step of performing feature area recognition processing on the target oral dental and jaw 3D scan model based on the surface curvature analysis method to construct a 3D tooth and gum segmentation model specifically includes: Scan the target oral cavity with a scanner to build a three-dimensional model of the jaw; Obtain the average surface curvature of the three-dimensional dental model and set the curvature threshold; The parts of the surface with an average curvature greater than a curvature threshold in the three-dimensional model of the jaw are marked as feature areas and extracted to construct a three-dimensional model of tooth and gum segmentation.

4. The method for eliminating groove lines in a gingival segmentation model extracted from an oral scan model according to claim 3, characterized in that: The step of sequentially performing concave compensation and gap filling processing on the homogenized pseudo 3D model to obtain a pre-processed pseudo 3D model specifically includes: Perform data analysis on the point cloud data in the homogenized pseudo 3D model, set the number of detection distance points in the horizontal and vertical directions, obtain and mark the sparse part of the point cloud data, and determine the point cloud data to be compensated; Through the expansion algorithm, a new point cloud is inserted into the point cloud data to be compensated according to the target point cloud density to obtain a compensated pseudo 3D model; Scanning the compensated pseudo three-dimensional model to obtain gaps in the compensated pseudo three-dimensional model and extracting all endpoints at both ends of all gaps; Establish key point pairs at both ends of all endpoints through spatial position relationship matching algorithm; The key point pairs are continuously patched using a patching algorithm to obtain a pre-processed pseudo three-dimensional model.

5. The method for eliminating groove lines in a gingival segmentation model extracted from an oral scan model according to claim 4, characterized in that: The step of sequentially filling the regions of the pre-processed pseudo 3D model using the Flood Fill algorithm to obtain a plurality of region blocks with identifiers specifically includes: Obtaining the closed area of the preprocessed pseudo three-dimensional model; Based on the Flood Fill algorithm, the starting point and filling threshold are determined, and the multiple closed areas of the pre-processed pseudo 3D model are filled to obtain several area blocks; Marking a number of area blocks to obtain a number of area blocks with identifiers.

6. The method for eliminating groove lines in a gingival segmentation model extracted from an oral scan model according to claim 5, characterized in that: The step of determining the tooth groove line features of the marked area blocks, performing merging, filling and tooth groove line elimination processing, and obtaining a curvature feature three-dimensional model after eliminating the tooth groove lines through inverse mapping specifically includes: Performing tooth groove line feature judgment on several marked area blocks and merging them to obtain merged area blocks; Filling the gaps in the merged area blocks by a gap filling method to obtain area blocks after eliminating the tooth groove lines; The area block after the groove line is eliminated is inversely mapped to the tooth and gum segmentation three-dimensional model to obtain the curvature feature three-dimensional model after the groove line is eliminated.

7. The method for eliminating groove lines in a gingival segmentation model extracted from an oral scan model according to claim 6, characterized in that: The step of judging the tooth groove line features of the marked area blocks and merging them to obtain the merged area blocks specifically includes: Obtaining the geometric center, maximum vertical coordinate value and minimum vertical coordinate value of a plurality of area blocks with identification; Determine the coordinate difference of each area block according to the maximum vertical coordinate value and the minimum vertical coordinate value; If the maximum vertical coordinate value and the minimum vertical coordinate value of the two area blocks overlap, obtain the length of the overlap and compare the coordinate difference between the two area blocks, and select the area block with the smallest coordinate difference as the reference area; Set the threshold feature of the inter-tooth overlap of the molar part of the teeth; The ratio data between the overlapping length and the coordinate difference of the reference area is calculated and judged with the threshold feature of the inter-tooth overlap of the molar part of the tooth. If the ratio data is greater than the threshold feature of the inter-tooth overlap of the molar part of the tooth, the two area blocks are merged to obtain a merged area block.

8. The method for eliminating groove lines in a gingival segmentation model extracted from an oral scan model according to claim 7, characterized in that: The step of inversely mapping the area block after the groove lines are eliminated to the tooth and gum segmentation three-dimensional model to obtain the curvature feature three-dimensional model after the groove lines are eliminated specifically includes: Deleting the tooth region identifier in the region block after the tooth groove line is eliminated to obtain a pre-processed region block; Based on the three-dimensional model of tooth and gum segmentation corresponding to the preprocessed area block, each triangular face in the three-dimensional model is scanned sequentially, the coordinate points of the triangular face are projected, and the triangular faces whose coordinate points fall in the area block are retained to obtain a three-dimensional model with curvature characteristics after eliminating the groove lines.

9. A system for eliminating tooth groove lines in a gingival segmentation model extracted from an oral scan model, characterized in that: Includes the following modules: The first module is used to construct a 3D model of tooth and gum segmentation based on the surface curvature analysis method and perform projection and point cloud processing to obtain a homogenized pseudo 3D model; The second module is used to perform concave compensation and gap filling processing on the homogenized pseudo 3D model in sequence to obtain a pre-processed pseudo 3D model; The third module is used to sequentially fill regions of the pre-processed pseudo 3D model using a Flood Fill algorithm to obtain a number of region blocks with identifiers; The fourth module is used to judge the groove line features of several marked area blocks, merge and fill them, and eliminate the groove lines. After inverse mapping, a three-dimensional model of the curvature features after eliminating the groove lines is obtained.

Images