Method, apparatus and electronic device for performing three-dimensional reconstruction of teeth

Abstract

In this embodiment, a method, an apparatus, and an electronic device for three-dimensional reconstruction of teeth are disclosed. The method uses a pre-trained tooth recognition model to identify teeth from an image to be processed. If teeth are recognized, a first image including the tooth area is output, the edge position of the teeth is determined from the tooth area to obtain a second image marked with the tooth edge area, the second image and the depth image are fused to obtain a fused image, and three-dimensional reconstruction of the teeth is performed based on the fused image. Thus, through the above method, a fused image including the precise position area of the teeth can be obtained, and three-dimensional reconstruction of the teeth can be performed according to the fused image, enabling the construction of a more precise three-dimensional model of the teeth.

Description

Technical Field

[0001] The present invention relates to the field of data processing, and in particular, to a method, an apparatus, and an electronic device for performing three-dimensional reconstruction of teeth. Background Art

[0002] In the process of analyzing and treating teeth, a three-dimensional model of teeth is usually required. In the prior art, teeth information is usually obtained by oral scanning, and a three-dimensional model of teeth is constructed through the scanning results.

[0003] However, in the prior art, when performing oral scanning, the target object is often not distinguished, and the scanning results will include tissues such as gums, lips, and mucous membranes. These information are redundant interference information. Therefore, in order to obtain a teeth model, the scanning personnel need to accurately scan the target. However, this method has high technical requirements for the scanning personnel, a high error rate, and the obtained teeth model is not ideal. Summary of the Invention

[0004] In view of this, embodiments of the present invention disclose a method, an apparatus, and an electronic device for performing three-dimensional reconstruction of teeth, which can construct a more accurate three-dimensional model of teeth.

[0005] Embodiments of the present invention disclose a method for performing three-dimensional reconstruction of teeth, including:

[0006] Inputting an image to be processed into a pre-trained teeth recognition model; the teeth recognition model is obtained by training a teeth model to be trained with an image marked with teeth parts as a training sample;

[0007] If teeth are recognized in the image to be processed, obtaining a first image including a teeth region;

[0008] Determining the edge position of the teeth from the teeth region to obtain a second image; the second image is an image to be processed marked with a teeth edge region;

[0009] Fusing the second image and a depth image to obtain a fused image; the depth image corresponds to the image to be processed. Performing three-dimensional reconstruction of the teeth based on the fused image to obtain a three-dimensional model of the teeth.

[0010] Optionally, the determining the edge position of the teeth from the teeth region to obtain a second image includes:

[0011] Performing smoothing processing on the first image to obtain a third image;

[0012] Calculating the gradient intensity and direction of each pixel point in the third image;

[0013] Screen the pixel points in the third image according to the gradient intensity and direction of each pixel;

[0014] Determine the first edge and the second edge in the third image by means of double-threshold detection;

[0015] Process the first edge and the second edge to obtain the edge positions of the teeth in the tooth region.

[0016] Optionally, the training process of the tooth recognition model includes:

[0017] Obtain training samples; the training samples are images marked with tooth parts;

[0018] Construct a convolutional neural network model to obtain a tooth recognition model to be trained;

[0019] Train the tooth recognition model to be trained with the training samples until convergence to obtain a tooth recognition model.

[0020] Optionally, the fusion of the second image and the depth image includes:

[0021] Determine the corresponding relationship of the pixels in the second image and the depth image;

[0022] Match the second image and the depth image based on the method of pixel matching and the corresponding relationship of each pixel in the second image and the depth image.

[0023] Optionally, the three-dimensional reconstruction of the teeth based on the fused image includes:

[0024] Obtain a point cloud model of the teeth from the fused image by the ORBSLAM2 method;

[0025] Reconstruct a surface model of the point cloud by the Poisson algorithm to obtain a three-dimensional model of the teeth.

[0026] An embodiment of the present invention discloses a device for three-dimensional reconstruction of teeth, including:

[0027] An input unit, configured to input an image to be processed into a pre-trained tooth recognition model; the tooth recognition model is obtained by training a tooth model to be trained with an image marked with tooth parts as training samples;

[0028] A recognition unit, configured to obtain a first image including a tooth region if teeth are recognized in the image to be processed;

[0029] An edge determination unit, configured to determine the edge positions of the teeth from the tooth region to obtain a second image; the second image is an image to be processed marked with a tooth edge region;

[0030] A fusion unit, configured to fuse the second image and the depth image to obtain a fused image; the depth image corresponds to the image to be processed.

[0031] A 3D reconstruction unit, configured to perform 3D reconstruction on the teeth based on the fused image to obtain a 3D model of the teeth.

[0032] Optionally, the edge determination unit includes:

[0033] A smoothing processing subunit, configured to perform smoothing processing on the first image to obtain a third image;

[0034] A calculation subunit, configured to calculate the gradient intensity and direction of each pixel point in the third image;

[0035] A screening subunit, configured to screen the pixel points in the third image according to the gradient intensity and direction of each pixel;

[0036] A first determination subunit, configured to determine a first edge and a second edge in the third image by means of double-threshold detection;

[0037] An edge determination subunit, configured to process the first edge and the second edge to obtain the edge positions of the teeth in the tooth region.

[0038] Optionally, it further includes:

[0039] An acquisition subunit, configured to acquire training samples; the training samples are images marked with tooth parts;

[0040] A construction subunit, configured to construct a convolutional neural network model to obtain a to-be-trained tooth recognition model;

[0041] A training subunit, configured to train the to-be-trained tooth recognition model by using the training samples until convergence to obtain a tooth recognition model.

[0042] Optionally, the fusion unit includes:

[0043] A second determination subunit, configured to determine the corresponding relationship between the pixels in the second image and the depth image;

[0044] A matching subunit, configured to match the second image and the depth image based on the method of pixel matching and the corresponding relationship between the respective pixels in the second image and the depth image.

[0045] An embodiment of the present invention discloses an electronic device, including:

[0046] A memory and a processor;

[0047] The memory is used to store a program, and the processor executes the method for three-dimensional tooth reconstruction described above when executing the program in the memory.

[0048] In this embodiment, a method, apparatus, and electronic device for three-dimensional tooth reconstruction are disclosed. The method uses a pre-trained tooth recognition model to identify teeth from an image to be processed. If teeth are recognized, a first image including the tooth region is output, the edge position of the teeth is determined from the tooth region to obtain a second image marked with the tooth edge region, the second image and the depth image are fused to obtain a fused image, and three-dimensional reconstruction of the teeth is performed based on the fused image. Thus, through the above method, a fused image including the precise position region of the teeth can be obtained, and three-dimensional reconstruction of the teeth can be performed based on the fused image. Without user-assisted scanning, a more precise three-dimensional model of the teeth can also be constructed. Description of the Drawings

[0049] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only the embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on the provided drawings.

[0050] Figure 1 A flowchart showing a method for three-dimensional tooth reconstruction provided by an embodiment of the present invention is shown;

[0051] Figure 2 A schematic diagram of a tooth region is shown;

[0052] Figure 3 A schematic diagram of the structure of an apparatus for three-dimensional tooth reconstruction disclosed by an embodiment of the present invention is shown;

[0053] Figure 4 A schematic diagram of the structure of an electronic device disclosed by an embodiment of the present invention is shown. Detailed Embodiments

[0054] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention.

[0055] In the prior art, since there may be redundant information in the oral scan data, a three-dimensional model of the teeth is obtained based on the scanned object. However, the obtained three-dimensional model may include redundant tissues such as gums and lips.

[0056] Therefore, in order to obtain a more accurate model of the teeth, this embodiment discloses a method for three-dimensional reconstruction of teeth. The method uses a pre-trained tooth recognition model to identify teeth from the image to be processed. If teeth are recognized, a first image including the tooth region is output, the edge position of the teeth is determined from the tooth region, and a second image marked with the tooth edge region is obtained. The second image and the depth image are fused to obtain a fused image, and three-dimensional reconstruction of the teeth is performed based on the fused image. Thus, through the above method, a fused image including the accurate position region of the teeth can be obtained, and three-dimensional reconstruction of the teeth is performed based on the fused image, so that an accurate tooth model can be obtained.

[0057] The following user names are explained:

[0058] ORBSLAM2 is a three-dimensional positioning and mapping algorithm (SLAM) based on ORB features. This algorithm was published by Raul Mur-Artal, J.M.M.Montiel, and Juan D. Tardos in IEEE Transactions on Robotics in 2015. Later, it was extended to Stereo and RGB-D sensors, which is called ORBSLAM2. ORB-SLAM is based on the PTAM architecture and adds functions of map initialization and loop closure detection.

[0059] Embodiment 1:

[0060] Reference Figure 1 , which shows a schematic flowchart of a method for three-dimensional reconstruction of teeth provided by an embodiment of the present invention. In this embodiment, the method includes:

[0061] S101: Input the image to be processed into a pre-trained tooth recognition model; the tooth recognition model is obtained by training a to-be-trained tooth model with an image marked with tooth parts as training samples.

[0062] In this embodiment, the image to be processed may be an image obtained by oral scanning, such as an image collected by an RGB-D camera.

[0063] In this embodiment, the tooth recognition model is obtained by training a to-be-trained tooth recognition model with training samples. The obtained tooth recognition model has the ability to recognize teeth from pictures.

[0064] Among them, the tooth recognition model can be any kind of convolutional neural network, or it can also be a combination of any kind of machine learning model, or it can also be a combination of multiple convolutional neural networks and a combination of multiple machine learning models. Optionally, the tooth recognition model can be a FastRCNN model.

[0065] In one implementation, the training process of the tooth recognition model includes:

[0066] Obtain training samples; the training samples are images marked with tooth parts;

[0067] Construct a convolutional neural network model to obtain a tooth recognition model to be trained;

[0068] Train the tooth recognition model to be trained with the training samples until convergence to obtain a tooth recognition model.

[0069] Furthermore, the trained tooth recognition model also needs to be verified through a validation set to check the accuracy of the tooth recognition model. When the accuracy is low, the tooth recognition model is trained again.

[0070] S102: If teeth are recognized in the image to be processed, obtain a first image containing the tooth region;

[0071] In this embodiment, the image to be processed may or may not contain teeth. The tooth recognition model has the ability to recognize teeth from the image to be processed. If the tooth recognition model recognizes that the image to be processed does not contain teeth, the image to be processed is discarded. If the tooth recognition model recognizes teeth from the image to be processed and determines the tooth region, a first image containing the tooth region can be output.

[0072] S103: Determine the edge position of the teeth from the tooth region to obtain a second image; the second image is the image to be processed marked with the tooth edge region;

[0073] In this embodiment, the tooth region obtained by the tooth recognition model is the region where the teeth are located. For example, it can be a rectangular region containing the teeth. For example Figure 2 As shown, the region of the rectangular frame in the figure can represent the tooth region recognized by the tooth recognition model.

[0074] However, the tooth region recognized by the tooth recognition model is only a general region and cannot accurately represent the position of the teeth. In order to obtain the accurate position of the teeth, in this embodiment, edge detection can be performed on the tooth region to obtain the accurate position of the teeth.

[0075] In this embodiment, the edge position of the tooth can be determined from the tooth region in various ways. Optionally, the following method can be used, and S103 includes:

[0076] Perform smoothing processing on the first image to obtain a third image;

[0077] Calculate the gradient intensity and direction of each pixel point in the third image;

[0078] Screen the pixel points in the third image according to the gradient intensity and direction of each pixel;

[0079] Determine the first edge and the second edge in the third image by means of double-threshold detection;

[0080] Process the first edge and the second edge to obtain the edge position of the tooth in the tooth region.

[0081] In this embodiment, in order to minimize the influence of noise on the edge detection result as much as possible, it is necessary to filter out the noise to prevent false detection caused by noise. To smooth the image, a Gaussian filter can be used to convolve with the image to achieve the smoothing processing of the first image and obtain a third image, so as to reduce the obvious noise influence on the edge detector. In this embodiment, the non-maximum suppression method can be adopted to screen the pixel points in the third image according to the gradient intensity and direction of each pixel.

[0082] In this embodiment, non-maximum suppression is an edge thinning technique, including:

[0083] 1) Compare the gradient intensity of the current pixel with two pixels along the positive and negative gradient directions.

[0084] 2) If the gradient intensity of the current pixel is the largest compared with the other two pixels, then this pixel point is retained as an edge point, otherwise this pixel point will be suppressed.

[0085] In this embodiment, after applying non-maximum suppression, the remaining pixels can more accurately represent the actual edges in the image. However, there are still some edge pixels caused by noise and color changes. To solve these stray responses, it is necessary to filter the edge pixels with weak gradient values and retain the edge pixels with high gradient values, which can be achieved by selecting high and low thresholds.

[0086] In this embodiment, in order to suppress the weak edges caused by noise, more accurate edge information can be obtained by suppressing isolated weak edges.

[0087] S104: Fuse the second image and the depth image to obtain a fused image; the depth image corresponds to the image to be processed; in this embodiment, the depth image corresponds to the image to be processed. For example, the depth image and the image to be processed are obtained by photographing the same target.

[0088] In this embodiment, there are various methods for fusing the second image and the depth image, which will not be elaborated herein. For example, the pixel point fusion method can be used to fuse the second image and the depth image. Specifically, S104 includes:

[0089] Determine the corresponding relationship of pixel points in the second image and the depth image;

[0090] Based on the pixel matching method and the corresponding relationship of each pixel point in the second image and the depth image, match the second image and the depth image.

[0091] In this embodiment, after fusing the second image and the depth image, the position of the teeth can be determined from the depth image.

[0092] S105: Perform three-dimensional reconstruction on the teeth based on the fused image to obtain a three-dimensional model of the teeth.

[0093] In this embodiment, there are various methods for performing three-dimensional reconstruction on the teeth according to the fused image, which are not limited in this embodiment. However, it should be noted that since the fused image contains the precise position of the teeth, an accurate three-dimensional model of the teeth can be obtained.

[0094] Optionally, in one implementation, the process of performing three-dimensional reconstruction on the teeth may include:

[0095] Obtain a point cloud model of the teeth from the fused image through the ORBSLAM2 method;

[0096] Perform surface model reconstruction on the point cloud through the Poisson algorithm to obtain a three-dimensional model of the teeth.

[0097] In this embodiment, the fused image is used as the input of the ORBSLAM2 method to estimate and track the pose of the camera, and the three-dimensional points of the key frames in the ORBSALM2 process are stitched to obtain a point cloud model of the teeth.

[0098] In this embodiment, teeth are recognized from the image to be processed by a pre-trained tooth recognition model. If teeth are recognized, a first image including the tooth region is output, the edge position of the teeth is determined from the tooth region, a second image marked with the tooth edge region is obtained, the second image and the depth image are fused to obtain a fused image, and three-dimensional reconstruction of the teeth is performed based on the fused image. Thus, a fused image including the accurate position region of the teeth can be obtained by the above method, and three-dimensional reconstruction of the teeth is performed according to the fused image, so that an accurate tooth model can be obtained.

[0099] Reference Figure 3 , shows a schematic structural diagram of an apparatus for three-dimensional reconstruction of teeth provided by an embodiment of the present invention. In this embodiment, the apparatus includes:

[0100] An input unit 301, configured to input the image to be processed into a pre-trained tooth recognition model; the tooth recognition model is obtained by training a to-be-trained tooth model with an image marked with tooth parts as training samples;

[0101] A recognition unit 302, configured to obtain a first image including the tooth region if teeth are recognized in the image to be processed;

[0102] An edge determination unit 303, configured to determine the edge position of the teeth from the tooth region to obtain a second image; the second image is the image to be processed marked with the tooth edge region;

[0103] A fusion unit 304, configured to fuse the second image and the depth image to obtain a fused image; the depth image corresponds to the image to be processed;

[0104] A three-dimensional reconstruction unit 305, configured to perform three-dimensional reconstruction of the teeth based on the fused image to obtain a three-dimensional model of the teeth.

[0105] Optionally, the edge determination unit includes:

[0106] A smoothing processing subunit, configured to perform smoothing processing on the first image to obtain a third image;

[0107] A calculation subunit, configured to calculate the gradient intensity and direction of each pixel point in the third image;

[0108] A screening subunit, configured to screen the pixel points in the third image according to the gradient intensity and direction of each pixel;

[0109] A first determination subunit, configured to determine a first edge and a second edge in the third image by a double-threshold detection method;

[0110] An edge determination subunit, configured to process the first edge and the second edge by suppressing isolated weak edges to obtain the edge positions of the teeth in the tooth region.

[0111] Optionally, it further includes:

[0112] An acquisition subunit, configured to acquire training samples; the training samples are images marked with tooth parts;

[0113] A construction subunit, configured to construct a convolutional neural network model to obtain a to-be-trained tooth recognition model;

[0114] A training subunit, configured to train the to-be-trained tooth recognition model with the training samples until convergence to obtain a tooth recognition model.

[0115] Optionally, the fusion unit includes:

[0116] A second determination subunit, configured to determine the correspondence of pixels in the second image and the depth image;

[0117] A matching subunit, configured to match the second image and the depth image based on the method of pixel matching and the correspondence of each pixel in the second image and the depth image.

[0118] An embodiment of the present invention discloses an electronic device, including:

[0119] A memory and a processor;

[0120] The memory is used to store a program, and the processor executes the method for performing three-dimensional reconstruction of teeth as described above when executing the program in the memory.

[0121] The device in this embodiment identifies teeth from the image to be processed through a pre-trained tooth recognition model. If teeth are recognized, it outputs a first image including the tooth region, determines the edge positions of the teeth from the tooth region to obtain a second image marked with the tooth edge region, fuses the second image and the depth image to obtain a fused image, and performs three-dimensional reconstruction of the teeth based on the fused image. Thus, through the above method, a fused image including the accurate position region of the teeth can be obtained, and three-dimensional reconstruction of the teeth can be performed according to the fused image. Without user-assisted scanning, a more accurate three-dimensional model of the teeth can also be constructed.

[0122] Reference Figure 4 , shows a schematic structural diagram of an electronic device disclosed in the present invention. In this embodiment, the electronic device includes:

[0123] A memory 401 and a processor 402;

[0124] The memory 401 is used to store programs, and the processor 402 executes the method for three-dimensional tooth reconstruction described below when executing the programs in the memory:

[0125] Input the image to be processed into a pre-trained tooth recognition model; the tooth recognition model is obtained by training a to-be-trained tooth model with images marked with tooth parts as training samples;

[0126] If teeth are recognized in the image to be processed, obtain a first image including the tooth area;

[0127] Determine the edge position of the teeth from the tooth area to obtain a second image; the second image is the image to be processed marked with the tooth edge area;

[0128] Fuse the second image and the depth image to obtain a fused image; the depth image corresponds to the image to be processed. Perform three-dimensional reconstruction of the teeth based on the fused image to obtain a three-dimensional model of the teeth.

[0129] Optionally, the determining the edge position of the teeth from the tooth area to obtain a second image includes:

[0130] Perform smoothing processing on the first image using a Gaussian filter to obtain a third image;

[0131] Calculate the gradient intensity and direction of each pixel point in the third image;

[0132] Use the maximum suppression method and the gradient intensity and direction of each pixel to screen the pixel points in the third image;

[0133] Determine the first edge and the second edge in the third image by the method of double-threshold detection;

[0134] Process the first edge and the second edge by the method of suppressing isolated weak edges to obtain the edge position of the teeth in the tooth area.

[0135] Optionally, the training process of the tooth recognition model includes:

[0136] Obtain training samples; the training samples are images marked with tooth parts;

[0137] Construct a convolutional neural network model to obtain a to-be-trained tooth recognition model;

[0138] Train the to-be-trained tooth recognition model with the training samples until convergence to obtain the tooth recognition model.

[0139] Optionally, the fusing of the second image and the depth image includes:

[0140] Determine the correspondence of pixels in the second image and the depth image;

[0141] Match the second image and the depth image based on the pixel matching method and the correspondence of each pixel in the second image and the depth image.

[0142] Optionally, the three-dimensional reconstruction of the teeth based on the fused image includes:

[0143] Obtain the point cloud model of the teeth from the fused image by the ORBSLAM2 method;

[0144] Perform surface model reconstruction on the point cloud by the Poisson algorithm to obtain the three-dimensional model of the teeth.

[0145] It should be noted that the various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The same or similar parts among the various embodiments can be referred to each other.

[0146] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for three-dimensional reconstruction of teeth, characterized in that, Including: Input the image to be processed into a pre-trained tooth recognition model; The tooth recognition model is obtained by training a tooth model to be trained with images marked with tooth parts as training samples; If teeth are recognized in the image to be processed, obtain a first image containing the tooth area; Determine the edge position of the teeth from the tooth area to obtain a second image; The second image is the image to be processed marked with the tooth edge area; Determine the corresponding relationship between the pixels in the second image and the depth image. Based on the pixel matching method and the corresponding relationship between each pixel in the second image and the depth image, match and fuse the second image and the depth image to obtain a fused image; the depth image corresponds to the image to be processed; Obtain the point cloud model of the teeth from the fused image through the ORBSLAM2 method, and reconstruct the surface model of the point cloud through the Poisson algorithm to obtain the three-dimensional model of the teeth.

2. The method according to claim 1, wherein The determining the edge position of the teeth from the tooth area to obtain a second image includes: Perform smoothing processing on the first image to obtain a third image; Calculate the gradient intensity and direction of each pixel point in the third image; Screen the pixel points in the third image according to the gradient intensity and direction of each pixel; Determine the first edge and the second edge in the third image through the double-threshold detection method; Process the first edge and the second edge to obtain the edge position of the teeth in the tooth area.

3. The method according to claim 1, wherein The training process of the tooth recognition model includes: Obtain training samples; the training samples are images marked with tooth parts; Construct a convolutional neural network model to obtain a tooth recognition model to be trained; Train the tooth recognition model to be trained with the training samples until convergence to obtain a tooth recognition model.

4. A device for three-dimensional reconstruction of teeth, characterized in that, Including: An input unit for inputting the image to be processed into a pre-trained tooth recognition model; The tooth recognition model is obtained by training a tooth model to be trained with images marked with tooth parts as training samples; A recognition unit for obtaining a first image containing the tooth area if teeth are recognized in the image to be processed; An edge determination unit for determining the edge position of the teeth from the tooth area to obtain a second image; The second image is the image to be processed marked with the tooth edge area; The fusion unit includes a second determination subunit and a matching subunit; The second determination subunit is used to determine the corresponding relationship between the pixels in the second image and the depth image; The matching subunit is used to match and fuse the second image and the depth image based on the pixel matching method and the corresponding relationship between each pixel in the second image and the depth image to obtain a fused image; the depth image corresponds to the image to be processed; A three-dimensional reconstruction unit for obtaining the point cloud model of the teeth from the fused image through the ORBSLAM2 method, and reconstructing the surface model of the point cloud through the Poisson algorithm to obtain the three-dimensional model of the teeth.

5. The device according to claim 4, characterized in that, The edge determination unit includes: A smoothing subunit, configured to perform smoothing processing on the first image to obtain a third image; A calculation subunit, configured to calculate the gradient intensity and direction of each pixel point in the third image; A screening subunit, configured to screen the pixel points in the third image according to the gradient intensity and direction of each pixel; A first determination subunit, configured to determine a first edge and a second edge in the third image by means of double-threshold detection; An edge determination subunit, configured to process the first edge and the second edge to obtain the edge positions of the teeth in the tooth region.

6. The device according to claim 4, characterized in that, The apparatus further includes: An acquisition subunit, configured to acquire a training sample; the training sample is an image marked with a tooth part; A construction subunit, configured to construct a convolutional neural network model to obtain a to-be-trained tooth recognition model; A training subunit, configured to train the to-be-trained tooth recognition model by using the training sample until convergence to obtain a tooth recognition model.

7. An electronic device, characterized in that, including: a memory and a processor; The memory is configured to store a program, and the processor, when executing the program stored in the memory, executes the method for three-dimensional tooth reconstruction according to any one of claims 1-3.

Images