A method, device, equipment and storage medium for generating dental crown data

Abstract

The present invention discloses a method, device, equipment and storage medium for generating crown data. A method for generating crown data includes: acquiring oral point cloud data, identifying abutment point cloud data, tooth point cloud data and background point cloud data based on the oral point cloud data; generating an occlusal projection map and an upper and lower jaw projection map based on the abutment point cloud data and the tooth point cloud data; processing the occlusal projection map and the upper and lower jaw projection map through a pre-trained crown generation model to obtain a crown image corresponding to the abutment point cloud data; and performing three-dimensional crown reconstruction based on the crown image to obtain a target crown model. The generation efficiency and accuracy of the crown are improved.

Description

Technical Field

[0001] The present invention relates to the field of oral technology, and in particular to a method, device, equipment and storage medium for generating crown data. Background Art

[0002] In the process of tooth restoration, the reasonable design of the crown is crucial to restore the function and beauty of the tooth. The correctness of occlusion is the basis for the tooth to perform its function, and the reconstruction of the correct occlusal contact relationship and the morphological characteristics of natural teeth is the key to the success of tooth restoration. Traditional crown generation technology mainly relies on manual production and a series of complex process flows. These methods include the wax model method, which has limitations in achieving precise occlusal design, and are overly dependent on the technician's operating experience and are highly subjective.

[0003] With the development of digital technology, computer-aided design technology has shown great potential in single crown design. Technicians design crown 3D models in a digital environment by operating 3D design software. This not only reduces the number of patient visits and physician operation time, but also improves the quality of restorations. However, existing computer-aided design solutions have high professional requirements for professionals and involve a large amount of user interaction, which to some extent affects the efficiency of crown 3D model generation. Summary of the invention

[0004] The present invention provides a method, device, equipment and storage medium for generating dental crown data, so as to solve the problems of low efficiency and low accuracy in generating dental crowns.

[0005] According to one aspect of the present invention, a method for generating dental crown data is provided, the method comprising:

[0006] Acquire oral point cloud data, and identify abutment point cloud data, tooth point cloud data, and background point cloud data based on the oral point cloud data;

[0007] Generate occlusal projection diagrams and maxillofacial projection diagrams based on abutment point cloud data and tooth point cloud data;

[0008] The occlusal projection and maxillary and mandibular projection are processed by the pre-trained crown generation model to obtain the crown image corresponding to the abutment point cloud data;

[0009] The three-dimensional crown reconstruction is performed based on the crown image to obtain the target crown model.

[0010] According to another aspect of the present invention, a device for generating crown data is provided, the device comprising:

[0011] A module for acquiring abutment point cloud data, tooth point cloud data and background point cloud data, which is used to acquire oral point cloud data and identify abutment point cloud data, tooth point cloud data and background point cloud data based on the oral point cloud data;

[0012] An occlusal projection diagram and maxillofacial projection diagram generation module is used to generate an occlusal projection diagram and maxillofacial projection diagrams based on abutment point cloud data and tooth point cloud data;

[0013] A crown image generation module is used to process the occlusal projection image and the maxillary and mandibular projection images through a pre-trained crown generation model to obtain a crown image corresponding to the abutment point cloud data;

[0014] The target crown model reconstruction module performs three-dimensional crown reconstruction based on the crown image to obtain the target crown model.

[0015] According to another aspect of the present invention, there is provided an electronic device, the electronic device comprising:

[0016] at least one processor; and

[0017] a memory communicatively connected to at least one processor; wherein,

[0018] The memory stores a computer program that can be executed by at least one processor. The computer program is executed by at least one processor so that the at least one processor can execute the method for generating dental crown data provided by any embodiment of the present invention.

[0019] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for generating crown data provided by any embodiment of the present invention when executed.

[0020] The technical solution of the embodiment of the present invention obtains oral point cloud data, identifies abutment point cloud data, tooth point cloud data and background point cloud data based on the oral point cloud data; generates an occlusal projection map and an upper and lower jaw projection map based on the abutment point cloud data and the tooth point cloud data; processes the occlusal projection map and the upper and lower jaw projection map through a pre-trained crown generation model to obtain a crown image corresponding to the abutment point cloud data; performs three-dimensional crown reconstruction based on the crown image to obtain a target crown model, thereby solving the problems of low crown generation efficiency and low accuracy and improving the reconstruction efficiency and accuracy of the target crown model.

[0021] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0023] Figure 1 is a flow chart of a method for generating crown data provided by Embodiment 1 of the present invention;

[0024] Figure 2 is a structural schematic diagram of a tooth crown generation model provided by an embodiment of the present invention;

[0025] Figure 3 It is a schematic diagram of the structure of a discriminator of a tooth crown generation model provided by an embodiment of the present invention;

[0026] Figure 4 It is a structural schematic diagram of a dental crown data generating device provided in Embodiment 2 of the present invention;

[0027] Figure 5 It is a structural schematic diagram of an electronic device provided in Embodiment 3 of the present invention. DETAILED DESCRIPTION

[0028] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0029] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0030] Embodiment 1

[0031] Figure 1This is a flowchart of a method for generating crown data provided in the first embodiment of the present invention. This embodiment is applicable to the case of automatically generating crown data. The method can be executed by a crown data generating device. The crown data generating device can be implemented in the form of hardware and / or software. The crown data generating device can be configured in the electronic device provided in the embodiment of the present invention. The electronic device can be a server, a computer, a mobile terminal or a medical device, etc. For example, the mobile terminal can be a mobile phone, a tablet computer, etc. For example, the medical device can be a device equipped with a processor, such as an oral medical device, which can be equipped with a point cloud scanning device and / or an oral surgical instrument, etc. Figure 1 As shown, the method includes:

[0032] S110, acquiring oral point cloud data, and identifying abutment point cloud data, tooth point cloud data, and background point cloud data based on the oral point cloud data.

[0033] In this embodiment, the oral point cloud data may be data obtained by scanning the oral cavity through an intraoral scanning device, for example, it may be oral three-dimensional model data, which can be used in oral medicine fields such as orthodontics, implantation, and restoration. Optionally, the oral scanning device may be an oral scanner. The oral point cloud data includes point data of multiple nodes, and the nodes in the oral point cloud data may include nodes located in the abutment area, nodes located in the tooth area, and nodes located in the background area, wherein the background area may include but is not limited to the gum area and the gum area.

[0034] After acquiring the oral point cloud data, it is necessary to perform data preprocessing on the oral point cloud data, for example, one or more of alignment processing, centroid centering processing, scale normalization processing and resampling processing.

[0035] Optionally, after acquiring the oral point cloud data, one or more of the following items are also included: aligning the point data of multiple nodes in the oral point cloud data with a predefined tooth coordinate system; centering the point data of multiple nodes in the oral point cloud data; normalizing the scale of the point data of multiple nodes in the oral point cloud data; and resampling the point data of multiple nodes in the oral point cloud data.

[0036] Specifically, the point data of multiple nodes in the oral point cloud data are aligned with the predefined tooth coordinate system, which can be done by aligning the coordinate system of the point data of multiple nodes in the oral scan data with the predefined tooth coordinate system through an iterative closest point algorithm, wherein the predefined tooth coordinate system may include standard oral data, and the standard oral data includes the point data of multiple nodes. The specific process is as follows: obtaining the point data of a node among multiple nodes in the oral point cloud data; performing the closest point matching step, by finding the corresponding relationship between each node in the oral point cloud data and the closest point in the predefined tooth coordinate system, establishing a matching pair between the node in the oral point cloud data and the node in the predefined tooth coordinate system, and aligning the oral point cloud data in the predefined tooth coordinate system through the above matching pair. Specifically, the gradient descent method is used to calculate the error between the matching node pairs, and the optimal transformation parameters are determined to minimize the error; according to the calculated optimal transformation parameters, the position and posture of the point data of multiple nodes in the oral point cloud data are adjusted to better align with the predefined tooth coordinate system. This process is repeated in the iterative process until the preset convergence condition is reached to ensure the accuracy and stability of the registration result. Finally, the point data of multiple nodes in the registered oral point cloud data are output to achieve precise alignment with the predefined tooth coordinate system.

[0037] Specifically, the centroid processing of the point data of multiple nodes in the oral point cloud data can be performed by moving the centroid of the upper and lower jaws to the origin of the coordinate system. The specific process is as follows: Calculate the coordinate average value of all nodes in the oral point cloud data, expressed as where p i represents the three-dimensional coordinates of point i, and n represents the number of nodes in the oral point cloud data; the coordinates of each node are subtracted from the average coordinate value to achieve centroid centralization, that is, where p′ i is the coordinate vector centered around the centroid.

[0038] Specifically, the scale normalization processing is performed on the point data of multiple nodes in the oral point cloud data. The minimum and maximum values ​​of the Y axis are calculated for the point cloud data after centroiding, and the coordinate values ​​of each node are normalized so that they fall within the range of [-1, 1]. The formula is expressed as:

[0039]

[0040] Among them, p″ i is the normalized coordinate vector; min(·) represents the minimum function; max(·) represents the maximum function.

[0041] Specifically, the point data of multiple nodes in the oral point cloud data are resampled, and the farthest point sampling algorithm can be used to evenly distribute the point data and keep the number of point clouds fixed at N, thereby optimizing the distribution and quality of the point data of multiple nodes in the oral point cloud data. The specific process is as follows: randomly select a point from the point data of multiple nodes in the oral point cloud data as the starting point set, denoted as P0; calculate the shortest distance from each remaining point in the point data of multiple nodes in the oral point cloud data to the starting point set P0, and the distance is determined by calculating the Euclidean distance from the point to all points in the point set and taking the minimum value; among all unselected points, select a point farthest from the existing point set and add it to the point set. Repeat the distance calculation and farthest point selection steps until the size of the point set reaches the predetermined N points. Through resampling processing, the number of nodes obtained by resampling is reduced, the amount of calculation in the tooth position segmentation process is reduced, and the uniformity of node layout is obtained by optimizing resampling.

[0042] The abutment point cloud data may include oral point cloud data located in the abutment area, and the tooth point cloud data may include oral point cloud data located in the tooth area. Among them, the abutment can be understood as the tooth that serves as the base of the denture, that is, the tooth to be repaired. The morphology of the abutment is different from that of other normal teeth. Correspondingly, the tooth area includes multiple normal teeth. The oral point cloud data can be identified by a machine learning model or a deep learning model, for example, it may include one or more of the PointNet model, the PointTransformer model, etc., to divide the oral point cloud data into abutment point cloud data and tooth point cloud data.

[0043] Optionally, abutment point cloud data, tooth point cloud data and background point cloud data are identified based on oral point cloud data, including: classifying the oral point cloud data based on a pre-trained classification model to obtain point types corresponding to multiple point data in the oral point cloud data, the point types including abutment type, tooth type and background type; the point data corresponding to the abutment type forms abutment point cloud data, and the point data corresponding to the tooth type forms tooth point cloud data.

[0044] Specifically, the classification model may include a neural network model, which may be used to classify the oral point cloud data to obtain the point type of the point data in the oral point cloud data, wherein the point type includes abutment type, tooth type, and background type, the abutment point cloud data includes point data corresponding to the abutment type, and the tooth point cloud data includes point data corresponding to the tooth type. The classification model may output different identifiers to represent the abutment type, tooth type, and background type, respectively.

[0045] S120, generating an occlusal projection diagram and an upper and lower jaw projection diagram based on the abutment point cloud data and the tooth point cloud data.

[0046] In this embodiment, the occlusal projection map can characterize the occlusal depth of the occlusal point data in the abutment point cloud data and the tooth point cloud data when the upper and lower teeth are in occlusion. The present invention can determine the occlusal projection map in a static occlusal mode, and can obtain the occlusal projection map through the occlusal point data in the abutment point cloud data and the tooth point cloud data. The occlusal point data may include contact point data in the upper and lower tooth point cloud data in a static occlusal state; the occlusal point data may be determined based on the distance between the upper and lower tooth point cloud data.

[0047] The maxillary and mandibular projection images can characterize the morphology of the jaw plane, wherein the jaw plane may include an imaginary plane formed by the contact point of the central incisor and the mesiobuccal cusps of the bilateral first molars of the jaw. The maxillary and mandibular projection images can be obtained by projecting the abutment point cloud data and the tooth point cloud data on the jaw plane respectively. The maxillary and mandibular projection images may include a mandibular projection depth map and a maxillary projection depth map.

[0048] Optionally, the associated tooth point cloud data of the abutment tooth point cloud data are determined in the oral point cloud data, wherein the associated tooth point cloud data of the abutment tooth point cloud data include adjacent tooth point cloud data of the abutment tooth point cloud data, and opposing tooth point cloud data corresponding to the abutment tooth point cloud data and the adjacent tooth point cloud data respectively; the occlusal point set is determined based on the associated tooth point cloud data, including: for each item of adjacent tooth point cloud data, a plurality of occlusal point data are determined by calculating the distance between the adjacent tooth point cloud data and the corresponding point data in the opposing oral point cloud data, the plurality of occlusal point data form an occlusal point set, wherein the occlusal point set includes a plurality of occlusal point data in the associated tooth point cloud data; the plurality of occlusal point data in the occlusal point set are projected to obtain an occlusal projection diagram; and the upper and lower jaw projection diagrams are generated based on the depth data between the abutment tooth point cloud data and the associated tooth point cloud data and the first reference plane respectively.

[0049] Specifically, the associated teeth may represent teeth associated with the abutment teeth, and may include adjacent teeth, opposing teeth of the abutment teeth, and opposing teeth of the adjacent teeth. The associated tooth point cloud data may include tooth point cloud data corresponding to the associated teeth, for example, may include tooth point cloud data corresponding to the adjacent teeth and the opposing teeth, respectively. The bite point may characterize the contact points between the associated teeth. The bite point data may include point data corresponding to the contact points. The bite point set may include a data set consisting of multiple bite point data. For example, the bite point set may be obtained by the K nearest neighbor algorithm, and the K nearest neighbor algorithm is used to obtain the K contact points closest to the adjacent tooth point cloud data from multiple point data in the opposing tooth point cloud data corresponding to the adjacent tooth point cloud data, and the point data corresponding to the K contact points constitute the bite point set.

[0050] The depth data can characterize the distances between the abutment point cloud data and the associated tooth point cloud data and the first reference plane, respectively. The first reference plane can be used to determine the desired plane of the depth data corresponding to the abutment point cloud data and the depth data corresponding to the associated tooth point cloud data, for example, it can include the maxillary surface and the mandibular surface. The distance between the abutment point cloud data and the first reference plane and the distance between the associated tooth point cloud data and the first reference plane are calculated respectively, and the distance between the abutment point cloud data and the first reference plane and the distance between the associated tooth point cloud data and the first reference plane are converted into grayscale values ​​corresponding to the abutment point cloud data and the associated tooth point cloud data. The distance between the abutment point cloud data and the first reference plane and the distance between the associated tooth point cloud data and the first reference plane are converted into grayscale values ​​corresponding to the abutment point cloud data and the associated tooth point cloud data by linear conversion. The larger the distance between the abutment point cloud data and the first reference plane or the distance between the associated tooth point cloud data and the first reference plane, the smaller the grayscale value corresponding to the abutment point cloud data or the grayscale value corresponding to the associated tooth point cloud data. The upper and lower jaw projection images are generated by the grayscale values ​​corresponding to the abutment point cloud data and the associated tooth point cloud data. The adjacent teeth point cloud data may include tooth point cloud data corresponding to the adjacent teeth. The opposing jaw oral point cloud data may include tooth point cloud data corresponding to the opposing jaw teeth of the abutment teeth and tooth point cloud data corresponding to the opposing jaw teeth of the adjacent teeth. The adjacent teeth point cloud data and the opposing jaw oral point cloud data constitute the associated tooth point cloud data of the abutment tooth point cloud data.

[0051] S130, processing the occlusal projection image and the maxillary and mandibular projection images using the pre-trained crown generation model to obtain a crown image corresponding to the abutment point cloud data.

[0052] In this embodiment, the crown generation model may include a neural network model, which may be used to generate a crown image corresponding to the abutment point cloud data. Optionally, the crown generation model may include but is not limited to a Transformer model and a generative adversarial network model. The model structure of the crown generation model is not limited here and may be created according to user needs. The crown image may include the crown and associated teeth corresponding to the abutment point cloud data.

[0053] Optionally, the crown generation model includes an encoder and a decoder; the encoder includes a first convolution processing module and multiple stage processing modules; the decoder includes multiple upsampling fusion modules and a second convolution module; wherein each stage processing module includes a coding block and a multi-scale attention module.

[0054] Specifically, the crown generation model includes an encoder and a decoder. The encoder includes a first convolution processing module and a plurality of stage processing modules connected in sequence, and each stage processing module includes a coding block and a multi-scale attention module connected in sequence. The first convolution processing module is used to extract low-level texture features of the input image, and the plurality of stage processing modules are used to extract semantic features of different depths. The plurality of stage processing modules may include different structures or the same structure. The decoder may include a plurality of upsampling fusion modules and a second convolution module, the upsampling fusion module is used to perform upsampling, splicing and other operations on the features extracted by the plurality of stage processing modules in the encoder, and the second convolution module is used to process the spliced ​​features. The encoder may include a plurality of stage processing modules, and the number of the stage processing modules may be four or five, and may be set according to requirements, and is not limited here.

[0055] For example, see Figure 2 , Figure 2 It is a structural schematic diagram of a crown generation model provided by an embodiment of the present invention. The crown generation model includes an encoder and a decoder, wherein the encoder includes a first convolution processing module and a plurality of stage processing modules connected in sequence, wherein the first convolution processing module includes a Conv Stem module, and taking four stage processing modules as an example, the four stage processing modules are Res-Block1 module, Res-Block2 module, Res-Block3 module and Res-Block4 module. The Conv Stem module includes a convolution layer with a convolution kernel size of 7×7 and a step size of 1×1, a batch normalization layer, an activation function layer and a maximum pooling layer with a kernel of 2×2, a step size of 2×2 and a padding of 1×1, and the activation function layer uses a ReLU activation function. The Res-Block1 module, the Res-Block2 module, the Res-Block3 module and the Res-Block4 module may include different structures, or may include the same structure. When the structures of Res-Block1, Res-Block2, Res-Block3 and Res-Block4 are the same, each module includes an encoding block and a multi-scale attention module, where the encoding block includes three sequentially connected convolutional layers with a convolution kernel size of 3×3 and a stride of 1×1, a downsampling convolutional layer, a batch normalization layer and an activation function layer, and the activation function layer adopts the ReLU activation function.

[0056] In order to improve the constraint ability of mutual information, a multi-scale attention module is introduced in each stage processing module. The multi-scale attention module can include multiple convolution blocks with different void rates, fusion processing blocks, context modeling modules and context aggregation modules. Any stage processing module can extract semantic features of different scales, and input the semantic features into convolution blocks with different void rates in turn for feature extraction to obtain multi-scale features. In order to reduce the redundancy of multi-scale feature maps, the multi-scale feature maps are spliced, for example, they can be spliced ​​in the channel dimension, and the convolution block with a convolution kernel of 1 is used to select the dimension of the spliced ​​multi-scale features to obtain multi-scale splicing features. The calculation formulas for multi-scale features and multi-scale splicing features are as follows:

[0057]

[0058] Among them, F in Indicates semantic features; F in，d represents the output of the convolutional block with a dilation rate of d; represents the dilated convolution; d represents the dilation rate, which can be set to different values, such as d∈[1, 3, 5, 7], which can be set according to the needs and is not limited here; k represents the convolution kernel size; pad represents the padding of the convolution kernel, which can be set to different values, such as pad∈[1, 3, 5, 7], which can be set according to the needs and is not limited here; F represents the multi-scale splicing feature; f BN Indicates batch normalization; f ReLU represents the activation function; Represents an inter-channel splicing operation.

[0059] In order to obtain global context features, the context modeling module can be used to obtain attention weights and aggregate features at all positions to generate global context features. The calculation formula for weighted global context features is as follows:

[0060]

[0061] in, Represents the features of the multi-scale splicing feature map F at pixel i, Represents x i The attention weight, W conv represents the convolution with a convolution kernel of 1, a i Represents the weighted global context features output by the context modeling module.

[0062] Since the interdependence between convolutional feature channels is extremely important for the feature representation ability of semantic segmentation, the importance of each channel is calculated through the bottleneck conversion operation of the feature conversion module to strengthen the channel interaction of the features. The addition operation can be used to aggregate the global context features to each position. The calculation formula of the global context feature is as follows:

[0063] z i =x i +W conv2 ReLU(LN(W conv1 (a i )))

[0064] Among them, z i represents the global context feature output by the global context aggregation module; W conv1 and W conv2 It is represented as convolution with kernel 1, which represents dimensionality reduction and dimensionality increase respectively; LN(·) represents layer normalization layer; ReLU(·) represents activation function.

[0065] In order to reduce the information loss caused by upsampling the low-resolution feature map to the high-resolution feature map, the decoder may include multiple upsampling fusion modules and a second convolution module, wherein the upsampling fusion module may include a residual convolution block. The iterative process of the decoder is expressed as follows:

[0066]

[0067] Among them, s j represents the features extracted by the sampling fusion module at stage j; h j represents the features extracted by the stage processing module of stage j; f up (·) represents upsampling operation, Represents the inter-channel splicing operation. Among them, the inputs of different upsampling fusion modules are different. For example, the input of the upsampling fusion module in the last stage includes the semantic features extracted by the stage processing module in the last stage and the semantic features extracted by the stage processing module in the previous stage; the input of the upsampling fusion modules in other stages except the upsampling fusion module in the last stage includes the features extracted by the upsampling fusion module in the previous stage and the semantic features extracted by the stage processing module in the current stage.

[0068] Optionally, the method also includes: extracting features of the occlusal projection image and the maxillary and mandibular projection images through a first convolution processing module to obtain image texture features; processing the image texture features in sequence through multiple stage processing modules to obtain multiple stage context features; wherein, in each stage processing module, encoding the input data through a coding block to obtain coding features, extracting multi-scale feature information corresponding to the coding features through a multi-scale attention module, and constructing the context features of the stage based on the multi-scale feature information; wherein the input data of the coding block includes image texture features or context features output by the previous stage processing module; upsampling and fusion processing are performed on multiple stage context features in sequence through multiple upsampling fusion modules to obtain fusion features; processing the fusion features through a second convolution module to obtain a crown image.

[0069] Specifically, the image texture feature can be understood as the feature obtained by extracting the occlusal projection map and the maxillary and mandibular projection map through the first convolution processing module. The stage context feature can be understood as the global context feature obtained by extracting the image texture feature through the stage processing module. The encoding feature can be understood as the feature obtained by encoding the input data. The multi-scale feature information can be understood as the information obtained by extracting the encoding feature through the multi-scale attention module. The fusion feature can be understood as the feature obtained by upsampling and fusing the stage context feature through the upsampling fusion module.

[0070] For the first-stage processing module of the first stage, the first-stage processing module is connected to the first convolution processing module, and the input data of the first-stage processing module includes image texture features; for other stage processing modules other than the first-stage processing module, they are respectively connected to the previous-stage processing modules, and the input data of other stage processing modules include context features output by the previous-stage processing modules.

[0071] Exemplarily, the multiple stage processing modules include a first stage processing module, a second stage processing module and a third stage processing module. In fact, the first stage processing module, the second stage processing module and the third stage processing module are connected in sequence, the image texture feature is input into the first stage processing module for feature extraction, and the first stage context feature is obtained. The first stage context feature is input into the second stage processing module for feature extraction to obtain the second stage context feature. The second stage context feature is input into the third stage processing module for feature extraction to obtain the third stage context feature. The number of stage processing modules can be three or five, and can be set according to needs, and there is no limitation here.

[0072] Taking N upsampling fusion modules as an example, the input of the Nth upsampling fusion module is the context feature of the Nth stage, the context feature of the Nth stage is upsampled to obtain the upsampled feature of the context feature of the Nth stage, the upsampled feature of the context feature of the Nth stage is fused with the context feature of the N-1th stage, for example, it can be a splicing process in the channel dimension to obtain the fusion feature of the Nth upsampling fusion module; the output feature of the Nth upsampling fusion module is upsampled to obtain the upsampled feature of the N-1th upsampling fusion module, the upsampled feature of the N-1th upsampling fusion module is fused with the fusion feature of the N-2th stage The processing may be, for example, a splicing process of the channel dimension to obtain the output features of the N-1th upsampling and fusion module; the output features of the N-1th upsampling and fusion module are upsampled to obtain the upsampling features of the N-2th upsampling and fusion module; the upsampling features of the N-2th upsampling and fusion module are spliced ​​with the N-3th stage fusion features, for example, a splicing process of the channel dimension to obtain the output features of the N-2th upsampling and fusion module; and so on to obtain the output features of the first upsampling and fusion module; the output features of the first upsampling and fusion module are processed through the second convolution module to obtain the crown image, thereby realizing the generation of the crown image.

[0073] In order to obtain a trained crown generation model, the untrained crown generation model can be trained, and the model parameters of the crown generation model can be adjusted to obtain a trained crown generation model.

[0074] Optionally, the training method of the crown generation model includes: obtaining a generator and a discriminator, the discriminator includes a global judgment module, a local judgment module and a fusion module, and the global judgment module and the local judgment module are respectively connected to the fusion module; obtaining sample point cloud data, and generating an occlusal projection map and an upper and lower jaw projection map corresponding to the sample point cloud data based on the sample point cloud data; inputting the occlusal projection map and the upper and lower jaw projection map corresponding to the sample point cloud data into the generator to obtain a crown image corresponding to the sample point cloud data, wherein the crown image includes a crown area and a tooth area; intercepting a crown sub-image corresponding to the crown area in the crown image; inputting the crown sub-image into the local judgment module of the discriminator, and inputting the crown image into the global judgment module of the discriminator to obtain discriminant data output by the fusion module; the generator takes minimizing the loss function as the goal, and the discriminator takes maximizing the loss function as the goal, and adopts an alternating training strategy to train the generator and the discriminator until a trained generator and discriminator are obtained, and the trained generator is determined as the crown generation model.

[0075] Specifically, the generator may include a neural network model, including but not limited to a Transformer model and a convolutional neural network model, which can be used to generate a crown image. Optionally, the generator includes an encoder and a decoder. The discriminator may include a neural network model, which can be used to optimize the generator. The sample point cloud data can be used to train the crown generation model. The discriminant data can be used to determine whether the input crown image is real, for example, it can be a probability value that the input crown image is a real image.

[0076] The discriminator includes a global judgment module, a local judgment module and a fusion module, and the global judgment module and the local judgment module are respectively connected to the fusion module. Among them, the global judgment module can be used to judge whether the input crown image is real, and the local judgment module can be used to judge whether the input crown sub-image is real. The fusion module can be used to fuse the output of the global judgment module and the output of the local judgment module, for example, splicing in the channel dimension can be performed. For example, see Figure 3 , Figure 3 : This is a schematic diagram of the structure of a discriminator of a crown generation model provided by an embodiment of the present invention. The global judgment module includes a 5-stage feature extractor, each of which includes a 3×3 convolution layer, a ReLU activation function layer, a batch normalization layer, and a 2×2 maximum pooling layer. The input of the global judgment module is the crown image F g and the first standard graph F t , the image size can be 256×256. The local judgment module includes four stages of feature extractors, each of which contains a 3×3 convolution layer, a ReLU activation function layer, a batch normalization layer, and a 2×2 maximum pooling layer. The input of the local judgment module is the crown sub-image F g ′ and the second standard graph F t′, the image size can be 128×128. The global determination module and the local determination module can have different numbers of feature extractors, or the same number of feature extractors, which can be set according to the needs, and there is no limitation here. Among them, the first standard image can be understood as the desired crown image, and the second standard image can be understood as the crown sub-image in the first standard image. When the global determination module and the local determination module have the same number of feature extractors, the input image of the global determination module and the input image of the local determination module have the same size; when the global determination module and the local determination module have different numbers of feature extractors, the input image size of the global determination module and the input image size of the local determination module are different. The fusion module fuses the output of the global determination module and the output of the local determination module, for example, the channel dimension can be spliced, and the fused data is passed through a linear layer to obtain the discriminant data. The global determination module can judge the crown image in the global dimension, which improves the accuracy of the crown image generated in the global dimension. The local determination module can judge the crown sub-image in the local dimension, which improves the accuracy of the crown sub-image generated in details. The output of the global judgment module and the output of the local judgment module are fused to improve the accuracy of the generation of crown images in terms of global dimensions and details, providing accurate data support for subsequent processing.

[0077] In some embodiments, the generator aims to minimize the loss function, and the discriminator aims to maximize the loss function. An alternating training strategy is used to train the generator and the discriminator. For example, a gradient descent algorithm can be used to train the generator and the discriminator until a trained generator and discriminator are obtained, and the trained generator is determined as a crown generation model.

[0078] S140, performing three-dimensional crown reconstruction based on the crown image to obtain a target crown model.

[0079] In this embodiment, the target crown model can be understood as the crown model corresponding to the crown image. The target crown model can be obtained by performing three-dimensional crown reconstruction on the crown image using a three-dimensional reconstruction algorithm.

[0080] Optionally, the tooth crown image is a grayscale image, and the grayscale value of a pixel in the grayscale image represents the distance between the point data corresponding to the pixel and the second reference plane.

[0081] Specifically, in the process of three-dimensional crown reconstruction, there is a virtual camera, and the plane where the virtual camera is located is used as the second reference plane. The grayscale value of the pixel point in the grayscale image can be converted into the distance between the point data corresponding to the pixel point and the second reference plane. For example, the grayscale value of the pixel point in the grayscale image can be converted into the distance between the point data corresponding to the pixel point and the second reference plane through a linear transformation. The first reference plane and the second reference plane can be different or the same. It can be understood that the first reference plane and the second reference plane are virtual planes.

[0082] Optionally, based on the grayscale value of each pixel in the crown image, the position information of the point data corresponding to the pixel is determined; the position information of the point data corresponding to each pixel in the crown image forms crown point cloud data; and a three-dimensional model is reconstructed based on the crown point cloud data to obtain a target crown model.

[0083] Specifically, the position information can characterize the specific position of the pixel point corresponding point data. For example, the position information may include two-dimensional coordinate information of the pixel point corresponding point data. According to the position information of the pixel point corresponding point data and the distance between the pixel point corresponding point data and the second reference plane, the pixel point corresponding point data is converted into crown point cloud data, and a three-dimensional model is reconstructed based on the crown point cloud data. For example, the crown point cloud data can be reconstructed by a three-dimensional reconstruction algorithm to obtain a target crown model. The target crown model may include a triangular mesh model.

[0084] The technical solution of this embodiment obtains oral point cloud data, identifies abutment point cloud data, tooth point cloud data and background point cloud data based on the oral point cloud data; generates an occlusal projection map and maxillary and mandibular projection maps based on the abutment point cloud data and the oral point cloud data; processes the occlusal projection map and maxillary and mandibular projection maps through a pre-trained crown generation model to obtain a crown image corresponding to the abutment point cloud data; performs three-dimensional crown reconstruction based on the crown image to obtain a target crown model, thereby improving the reconstruction efficiency and accuracy of the target crown model.

[0085] Embodiment 2

[0086] Figure 4 Schematic diagram of the structure of a crown data generating device provided by the second embodiment of the present invention. Figure 4 As shown, the device comprises:

[0087] Abutment point cloud data, tooth point cloud data and background point cloud data acquisition module 210, used to acquire oral point cloud data, and identify abutment point cloud data, tooth point cloud data and background point cloud data based on the oral point cloud data;

[0088] An occlusal projection diagram and maxillofacial projection diagram generation module 220 is used to generate an occlusal projection diagram and maxillofacial projection diagrams based on abutment point cloud data and oral point cloud data;

[0089] The tooth crown image generation module 230 is used to process the occlusal projection image and the maxillary and mandibular projection images through the pre-trained tooth crown generation model to obtain the tooth crown image corresponding to the abutment point cloud data;

[0090] The target crown model reconstruction module 240 is used to perform three-dimensional crown reconstruction based on the crown image to obtain a target crown model.

[0091] On the basis of the above embodiment, optionally, the tooth point cloud data and oral point cloud data acquisition module 210 is also used to: classify the oral point cloud data based on a pre-trained classification model to obtain point types corresponding to multiple point data in the oral point cloud data, the point types including abutment type, tooth type and background type; the point data corresponding to the abutment type forms the abutment point cloud data, and the point data corresponding to the tooth type forms the oral point cloud data.

[0092] Optionally, the occlusal projection map and maxillofacial projection map generation module 220 is also used to: determine the tooth point cloud data associated with the abutment point cloud data in the oral point cloud data; determine the occlusal point set based on the abutment point cloud data and the associated tooth point cloud data, the occlusal point set including multiple occlusal point data in the abutment point cloud data and the associated tooth point cloud data; perform projection processing on the multiple occlusal point data in the occlusal point set to obtain an occlusal projection map; generate maxillofacial projection maps based on the depth data between the abutment point cloud data and the associated tooth point cloud data and the first reference plane.

[0093] Optionally, the occlusal projection map and maxillofacial projection map generation module 220 is also used to: for each item of abutment point cloud data and adjacent tooth point cloud data, determine multiple occlusal point data based on the distance between the corresponding midpoint data of the opposing jaw oral point cloud data, and the multiple occlusal point data form an occlusal point set.

[0094] Optionally, the crown image generation module 230 is also used to: perform feature extraction on the occlusal projection image and the maxillary and mandibular projection images through the first convolution processing module to obtain image texture features; process the image texture features in sequence through multiple stage processing modules to obtain multiple stage context features; wherein, in each stage processing module, the input data is encoded by the encoding block to obtain the encoding features, the multi-scale feature information corresponding to the encoding features is extracted by the multi-scale attention module, and the context features of the stage are constructed based on the multi-scale feature information; wherein the input data of the encoding block includes image texture features or the context features output by the previous stage processing module; the multiple stage context features are upsampled and fused in sequence through multiple upsampling fusion modules to obtain fused features; the fused features are processed by the second convolution module to obtain a crown image.

[0095] Optionally, the crown image generation module 230 is also used for: training a crown generation model, and the training method of the crown generation model includes: obtaining a generator and a discriminator, the discriminator includes a global judgment module, a local judgment module and a fusion module, and the global judgment module and the local judgment module are respectively connected to the fusion module; obtaining sample point cloud data, and generating an occlusal projection map and an upper and lower jaw projection map corresponding to the sample point cloud data based on the sample point cloud data; inputting the occlusal projection map and the upper and lower jaw projection map corresponding to the sample point cloud data into the generator, and obtaining the occlusal projection map corresponding to the sample point cloud data. A crown image, which includes a crown area and a tooth area; intercepting a crown sub-image corresponding to the crown area in the crown image; inputting the crown sub-image into the local judgment module of the discriminator, and inputting the crown image into the global judgment module of the discriminator to obtain the discrimination data output by the fusion module; the generator takes minimizing the loss function as the goal, and the discriminator takes maximizing the loss function as the goal, adopts an alternating training strategy, trains the generator and the discriminator, until the trained generator and discriminator are obtained, and the trained generator is determined as the crown generation model.

[0096] Optionally, the target crown model reconstruction module 240 is also used to: determine the position information of the point data corresponding to the pixel point based on the grayscale value of each pixel point in the crown image; form crown point cloud data based on the position information of the point data corresponding to each pixel point in the crown image; and perform three-dimensional model reconstruction based on the crown point cloud data to obtain the target crown model.

[0097] The dental crown data generating device provided in the embodiment of the present invention can execute the dental crown data generating method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0098] Embodiment 3

[0099] Figure 5 1 is a schematic diagram of the structure of an electronic device provided in Embodiment 3 of the present invention. The electronic device 10 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0100] like Figure 5As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0101] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0102] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as a crown data generation method.

[0103] In some embodiments, the crown data generation method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the crown data generation method described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to execute the crown data generation method in any other appropriate manner (e.g., by means of firmware).

[0104] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0105] The computer program for implementing the method for generating dental crown data of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general-purpose computer, a special-purpose computer or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program can be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0106] Embodiment 5

[0107] Embodiment 5 of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to execute a method for generating crown data, the method comprising:

[0108] Acquire oral point cloud data, and identify abutment point cloud data, tooth point cloud data and background point cloud data based on the oral point cloud data; generate occlusal projection diagrams and maxillofacial projection diagrams based on the abutment point cloud data and oral point cloud data; process the occlusal projection diagrams and maxillofacial projection diagrams through a pre-trained crown generation model to obtain a crown image corresponding to the abutment point cloud data; perform three-dimensional crown reconstruction based on the crown image to obtain a target crown model.

[0109] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0110] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0111] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0112] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0113] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0114] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for generating crown data, characterized in that: include: Acquire oral point cloud data, and identify abutment point cloud data, tooth point cloud data, and background point cloud data based on the oral point cloud data; Generate an occlusal projection diagram and an upper and lower jaw projection diagram based on the abutment point cloud data and the tooth point cloud data; The occlusal projection image and the maxillary and mandibular projection images are processed by a pre-trained crown generation model to obtain a crown image corresponding to the abutment point cloud data; A three-dimensional dental crown reconstruction is performed based on the dental crown image to obtain a target dental crown model.

2. The method according to claim 1, characterized in that The identifying of abutment point cloud data, tooth point cloud data and background point cloud data based on the oral point cloud data includes: Classify the oral point cloud data based on a pre-trained classification model to obtain point types corresponding to a plurality of point data in the oral point cloud data, wherein the point types include abutment types, tooth types, and background types; The point data corresponding to the abutment type forms the abutment point cloud data, and the point data corresponding to the tooth type forms the tooth point cloud data.

3. The method according to claim 1, characterized in that The generating of the occlusal projection diagram and the maxillofacial projection diagram based on the abutment point cloud data and the tooth point cloud data comprises: Determining the tooth point cloud data associated with the abutment point cloud data in the tooth point cloud data; Determine a bite point set based on the associated tooth point cloud data, wherein the bite point set includes a plurality of bite point data in the associated tooth point cloud data; Projecting the plurality of occlusal point data in the occlusal point set to obtain the occlusal projection diagram; The maxillomandibular projection images are generated based on the depth data between the abutment point cloud data and the associated tooth point cloud data and the first reference plane.

4. The method according to claim 3, characterized in that The tooth point cloud data associated with the abutment point cloud data includes the tooth point cloud data adjacent to the abutment point cloud data, and the tooth point cloud data of the opposing jaws corresponding to the abutment point cloud data and the adjacent tooth point cloud data respectively; The determining of the occlusal point set based on the abutment point cloud data and the associated tooth point cloud data comprises: For each item of the adjacent tooth point cloud data, the multiple bite point data are determined based on the distance between the adjacent tooth point cloud data and the corresponding point data of the opposing tooth point cloud data, and the multiple bite point data form the bite point set.

5. The method according to claim 1, characterized in that The tooth crown image is a grayscale image, and the grayscale value of a pixel point in the grayscale image represents the distance between the corresponding point data of the pixel point and the second reference plane; The three-dimensional crown reconstruction based on the crown image to obtain a target crown model includes: Based on the gray value of each pixel in the dental crown image, position information of the point data corresponding to the pixel is determined; the position information of the point data corresponding to each pixel in the dental crown image forms dental crown point cloud data; A three-dimensional model is reconstructed based on the tooth crown point cloud data to obtain the target tooth crown model.

6. The method according to claim 1, characterized in that The crown generation model includes an encoder and a decoder; the encoder includes a first convolution processing module and multiple stage processing modules; the decoder includes multiple upsampling fusion modules and a second convolution module; Among them, each stage processing module includes an encoding block and a multi-scale attention module; The method further comprises: Extracting features of the occlusal projection image and the maxillofacial projection image through the first convolution processing module to obtain image texture features; The image texture features are processed in sequence by the multiple stage processing modules to obtain multiple stage context features; wherein, in each of the stage processing modules, the input data is encoded by the encoding block to obtain the encoding feature, the multi-scale feature information corresponding to the encoding feature is extracted by the multi-scale attention module, and the context feature of the stage is constructed based on the multi-scale feature information; wherein, the input data of the encoding block includes the image texture feature or the context feature output by the previous stage processing module; The multiple upsampling and fusion modules are used to sequentially perform upsampling and fusion processing on the multiple stage context features to obtain fusion features; The fusion features are processed by the second convolution module to obtain the tooth crown image.

7. The method according to claim 1 or 6, characterized in that: The training method of the tooth crown generation model comprises: Acquire a generator and a discriminator, wherein the discriminator includes a global determination module, a local determination module and a fusion module, and the global determination module and the local determination module are respectively connected to the fusion module; Acquire sample point cloud data, and generate an occlusal projection diagram and an upper and lower jaw projection diagram corresponding to the sample point cloud data based on the sample point cloud data; Inputting the occlusal projection image and the maxillomandibular projection image corresponding to the sample point cloud data into the generator to obtain a crown image corresponding to the sample point cloud data, wherein the crown image includes a crown area and a tooth area; intercepting a crown sub-image corresponding to the crown area in the crown image; inputting the crown sub-image into the local determination module of the discriminator, and inputting the crown image into the global determination module of the discriminator to obtain the determination data output by the fusion module; The generator aims to minimize the loss function, and the discriminator aims to maximize the loss function. An alternating training strategy is adopted to train the generator and the discriminator until a trained generator and discriminator are obtained, and the trained generator is determined as the crown generation model.

8. A dental crown data generating device, characterized in that: include: A module for acquiring abutment point cloud data, tooth point cloud data and background point cloud data, for acquiring oral point cloud data, and identifying abutment point cloud data, tooth point cloud data and background point cloud data based on the oral point cloud data; An occlusal projection diagram and maxillofacial projection diagram generation module, used for generating an occlusal projection diagram and maxillofacial projection diagrams based on the abutment point cloud data and the tooth point cloud data; A crown image generation module, used to process the occlusal projection image and the maxillary and mandibular projection images through a pre-trained crown generation model to obtain a crown image corresponding to the abutment point cloud data; The target crown model reconstruction module is used to perform three-dimensional crown reconstruction based on the crown image to obtain a target crown model.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for generating dental crown data according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for generating dental crown data according to any one of claims 1 to 7 when executed.

Images