Oral image processing method, device, electronic equipment and computer storage medium

Abstract

The present invention relates to an oral image processing method, device, electronic device and computer storage medium, the method comprising: obtaining an oral image to be processed, the oral image to be processed including a periapical film after root canal filling; binarizing the oral image to be processed to obtain a binary image; and evaluating the filling quality of the root canal filling area in the oral image to be processed according to the oral image to be processed and the binary image to obtain an evaluation result. By the method of the present invention, the oral image to be processed is converted into a binary image, and more detailed information of the root canal filling area can be reflected through the binary image. When evaluating the filling quality of the root canal filling area, the root canal filling area in the oral image to be processed and the binary image is comprehensively considered, so that the evaluation result obtained can be more accurate.

Description

Technical Field

[0001] The present invention relates to the field of image processing technology, and in particular to an oral image processing method, device, electronic equipment and computer storage medium. Background Art

[0002] Root canal therapy is a common and effective treatment for pulp disease and periapical periodontitis. The treatment process is complex and requires high technical requirements, and root canal filling is a key step in root canal therapy. Root canal filling refers to filling the root canal filling material into the prepared and disinfected root canal to eliminate dead space, isolate the communication between the root canal and the periapical tissue, and prevent reinfection in the root canal. It can be seen that good and complete root canal filling is the fundamental guarantee for achieving the expected therapeutic effect of root canal treatment. How to correctly evaluate the effect of root canal filling is crucial to do a good job in root canal quality control and improve the quality of root canal treatment.

[0003] The traditional root canal filling evaluation standard uses X-rays of the affected teeth to observe the distance from the root canal filling material to the apical foramen and the gap between the filling material and the root canal wall to judge the root canal filling effect. This is prone to defects such as image overlap, deformation, and artifacts. Moreover, such evaluation mainly relies on the patient's feelings during the operation and the doctor's subjective judgment, which is often not objective and accurate enough. Therefore, in the existing technology, how to automatically and effectively evaluate the quality of root canal filling is a technical problem that needs to be solved urgently. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide an oral image processing method, device, electronic equipment and computer storage medium, aiming to solve the problem of how to automatically and effectively evaluate the quality of root canal filling.

[0005] The technical solution of the present invention to solve the above technical problem is as follows: an oral image processing method, the method comprising:

[0006] Acquire an oral cavity image to be processed, wherein the oral cavity image to be processed includes a periapical film after root canal filling;

[0007] Binarization is performed on the oral cavity image to be processed to obtain a binary image;

[0008] According to the oral image to be processed and the binary image, the filling quality of the root canal filling area in the oral image to be processed is evaluated to obtain an evaluation result.

[0009] The beneficial effect of the present invention is that after the root canal is filled in the periapical film, the grayscale values ​​corresponding to the root canal filling area and the non-root canal filling area are different. After the oral image to be processed is binarized, the root canal filling area and the non-root canal filling area can be more clearly distinguished through the binary image. More detailed information of the root canal filling area can be reflected through the binary image. When evaluating the filling quality of the root canal filling area, the root canal filling area in the oral image to be processed and the root canal filling area in the binary image are comprehensively considered, so that the obtained evaluation result can be more accurate.

[0010] Based on the above technical solution, the present invention can also be improved as follows.

[0011] Furthermore, the method further comprises:

[0012] Cropping the binary image to obtain a first image including a root canal filling area;

[0013] Cropping the oral cavity image to be processed to obtain a second image including a root canal filling area;

[0014] The filling quality of the root canal filling area in the oral image to be processed is evaluated according to the oral image to be processed and the binary image, and the evaluation result is obtained, including:

[0015] The filling quality of the root canal filling area in the oral cavity image to be processed is evaluated according to the first image and the second image to obtain an evaluation result.

[0016] The beneficial effect of adopting the above further scheme is that since the oral image to be processed includes not only the image corresponding to the root canal filling area but also the image corresponding to the non-root canal filling area, in the scheme of the present invention, by cropping out the root canal filling area in the binary image and the root canal filling area in the oral image to be processed, the information of the non-root canal filling area in the binary image and the oral image to be processed can be filtered out, and the information of the non-root canal filling area is irrelevant to the filling quality of the root canal filling, thereby reducing the amount of data calculation and improving the accuracy of the evaluation results.

[0017] Furthermore, the above-mentioned cropping of the binary image to obtain a first image containing the root canal filling area includes:

[0018] According to the binary image, the position of the root canal filling area in the binary image is determined;

[0019] According to the position of the root canal filling area in the binary image, the binary image is cropped to obtain a first image including the root canal filling area;

[0020] The above-mentioned method of cropping the oral cavity image to be processed to obtain a second image containing the root canal filling area includes:

[0021] Determining a second position of the root canal filling area in the oral cavity image to be processed according to the first image;

[0022] The oral cavity image to be processed is cropped according to the second position to obtain a second image including the root canal filling area.

[0023] The beneficial effect of adopting the above further scheme is that since the first image and the second image are both images containing the root canal filling area in the oral image to be processed, when cropping the binary image, the position of the root canal filling area in the binary image can be determined first, and then the binary image can be cropped according to the position. When cropping the oral image to be processed, the second position of the root canal filling area in the oral image to be processed can be determined with reference to the first image, thereby ensuring that the image areas corresponding to the second image and the first image are both root canal filling areas.

[0024] Furthermore, the above method of cropping the oral cavity image to be processed according to the second position to obtain a second image including the root canal filling area includes:

[0025] According to the second position, generating a cropping bounding box at a position corresponding to the second position in the oral image to be processed;

[0026] Expand the cropping bounding box to obtain an expanded bounding box;

[0027] The oral cavity image to be processed is cropped according to the enlarged bounding box to obtain a second image including the root canal filling area.

[0028] The beneficial effect of adopting the above further scheme is that when the processed oral image is cropped according to the second position, the cropping can be performed according to the cropping bounding box determined by the second position. Since the cropping bounding box corresponds to the root canal filling area, considering that there may be some information around the root canal filling area that may affect the filling quality, in the scheme of the present invention, the cropping bounding box can be expanded. By using the expanded bounding box to crop the processed oral image, the cropped second image can include not only the root canal filling area, but also the information around the root canal filling area. Therefore, when the filling quality of the root canal filling area is evaluated based on the first image and the second image, the evaluation result is more accurate.

[0029] Furthermore, the filling quality of the root canal filling area in the oral image to be processed is evaluated based on the oral image to be processed and the binary image, and the evaluation result is obtained, including:

[0030] Extracting a first image feature of the oral image to be processed and a second image feature of the binary image respectively, wherein the first image feature is a regional feature of the root canal filling area, and the second image feature is an edge feature of the root canal filling area;

[0031] The filling quality of the root canal filling area in the oral cavity image to be processed is evaluated according to the first image feature and the second image feature to obtain an evaluation result.

[0032] The beneficial effect of adopting the above further scheme is that the filling features of different levels can be reflected by the first image feature and the second image feature, and the features at the edge of the root canal filling area can be reflected by the second image feature, so that when the filling quality of the root canal filling area in the oral image to be processed is evaluated based on the first image feature and the second image feature, the evaluation result is more accurate.

[0033] Furthermore, the above-mentioned processing of cropping the binary image to obtain the first image containing the root canal filling area is realized by a root canal filling segmentation model, wherein the input of the above-mentioned root canal filling segmentation model is a binary image containing the periapical film after the root canal filling, and the output is the first image containing the root canal filling area;

[0034] The above-mentioned processing process of evaluating the filling quality of the root canal filling area in the oral image to be processed based on the first image and the second image to obtain the evaluation result is realized by a root canal filling quality category recognition model, wherein the input of the above-mentioned root canal filling quality category recognition model is the first image and the second image, and the output is the evaluation result of the root canal filling quality corresponding to the oral image to be processed.

[0035] The beneficial effect of adopting the above further solution is that the processing process of cropping the binary image to obtain the first image can be implemented by a root canal filling segmentation model, wherein the root canal filling segmentation model is trained in the following way:

[0036] Acquire multiple sample images, each sample image including a periapical film after the root canal is filled;

[0037] Perform binary processing on each sample image to obtain a binary image, wherein each pixel point in each binary image corresponds to a segmentation and annotation result, and for each pixel point in each binary image, the segmentation and annotation result represents whether the pixel point belongs to a root canal filling area or a non-root canal filling area in the binary image;

[0038] The first initial model is trained according to each binary image until the loss value of the first initial model meets the first training end condition, and the first initial model corresponding to the first training end condition is used as the root canal filling segmentation model.

[0039] The input of the first initial model is each binary image, and the output is the predicted segmentation result corresponding to each binary image. The loss value of the first initial model represents the difference between the predicted segmentation result corresponding to each binary image and the segmentation annotation result.

[0040] The above-mentioned process of evaluating the filling quality of the root canal filling area in the oral image to be processed according to the first image and the second image to obtain the evaluation result is realized by a root canal filling quality category recognition model, wherein the above-mentioned root canal filling quality category recognition model is trained in the following manner:

[0041] Acquire multiple sample images, each sample image corresponds to a quality annotation result, and the quality annotation result is any one of just filling, underfilling or overfilling;

[0042] For each sample image, determine a first image and a second image corresponding to the sample image, and regard the first image and the second image corresponding to each sample image as a sample image pair;

[0043] According to each sample image pair, the second initial model is trained until the loss value of the second initial model meets the second training end condition, and the second initial model corresponding to the second training end condition is used as the root canal filling quality category recognition model;

[0044] Among them, the input of the above-mentioned second initial model is each sample image pair, and the output is the prediction quality assessment result corresponding to each sample image pair. The loss value of the second initial model represents the difference between the prediction quality assessment result corresponding to each sample image pair and the quality labeling result.

[0045] The beneficial effect of adopting the above further scheme is that the neural network model is used in the scheme of the present invention to achieve the purpose of automatically evaluating the filling quality of the root canal filling area, making the evaluation process more intelligent.

[0046] Furthermore, before the above-mentioned binarization processing is performed on the oral cavity image to be processed to obtain the binary image, it also includes:

[0047] The grayscale value of the oral cavity image to be processed is normalized to obtain a processed oral cavity image.

[0048] The beneficial effect of adopting the above further scheme is that before the oral image to be processed is binarized, the grayscale value of the oral image to be processed can be normalized to improve image contrast and enhance features in the image, thereby making the evaluation result more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for describing the embodiments of the present invention are briefly introduced below.

[0050] Figure 1 A schematic flow chart of an oral image processing method provided by one embodiment of the present invention;

[0051] Figure 2A schematic flow chart of another oral image processing method provided by an embodiment of the present invention;

[0052] Figure 3 A schematic diagram of the structure of an oral image processing device provided by one embodiment of the present invention;

[0053] Figure 4 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0054] The principles and features of the present invention are described below. The examples given are only used to explain the present invention and are not used to limit the scope of the present invention.

[0055] The technical solution of the present invention and how the technical solution of the present invention solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present invention will be described below in conjunction with the accompanying drawings.

[0056] The solution provided by the embodiment of the present invention can be applied to any application scenario that requires automatic evaluation of the filling quality of root canal filling. The solution provided by the embodiment of the present invention can be executed by any electronic device, for example, it can be a user's terminal device, and the terminal device can include at least one of the following: a smart phone, a tablet computer, a laptop computer, a desktop computer, etc.

[0057] The embodiment of the present invention provides a possible implementation method, such as Figure 1 As shown, a flowchart of an oral image processing method is provided. The solution can be executed by any electronic device, for example, a terminal device, or a terminal device and a server. For the convenience of description, the method provided by the embodiment of the present invention will be described below by taking the terminal device as the execution subject as an example. Figure 1 The method may include the following steps:

[0058] Step S110, obtaining an oral cavity image to be processed, wherein the oral cavity image to be processed includes a periapical film after root canal filling;

[0059] Step S120, performing binarization processing on the oral cavity image to be processed to obtain a binary image;

[0060] Step S130 , evaluating the filling quality of the root canal filling area in the oral image to be processed according to the oral image to be processed and the binary image, to obtain an evaluation result.

[0061] Through the method of the present invention, after the root canal is filled in the periapical film, the grayscale values ​​corresponding to the root canal filling area and the non-root canal filling area are different. After the oral image to be processed is binarized, the root canal filling area and the non-root canal filling area can be more clearly distinguished through the binary image. More detailed information of the root canal filling area can be reflected through the binary image. When evaluating the filling quality of the root canal filling area, the root canal filling area in the oral image to be processed and the root canal filling area in the binary image are comprehensively considered, so that the obtained evaluation result can be more accurate.

[0062] The solution of the present invention is further described below in conjunction with the following specific embodiments. In this embodiment, the oral image processing method may include the following steps:

[0063] Step S110, obtaining an oral cavity image to be processed, wherein the oral cavity image to be processed includes a periapical film after root canal filling.

[0064] Among them, the oral image to be processed refers to an image whose filling quality of the root canal filling area needs to be evaluated. The image can be an image obtained by taking X-rays of the affected tooth, or it can be an image acquired by other professional image acquisition equipment. The specific method of acquiring the oral image to be processed is not limited in the present invention, and all are within the protection scope of the present invention.

[0065] Step S120, binarizing the oral cavity image to be processed to obtain a binary image.

[0066] Among them, a binary image means that there are only two grayscale levels in the image, that is, the grayscale value of any pixel in the image is 0 or 255, representing black and white respectively. Therefore, the binary image obtained after binarization of the oral image to be processed has only two colors, black and white. The root canal filling area and the non-root canal filling area in the image are distinguished by the two colors of black and white. Compared with the oral image to be processed, the binary image can reflect more detailed features of the root canal filling area.

[0067] Optionally, when binarizing the oral image to be processed, the grayscale value of each pixel corresponding to the root canal filling material in the image may be set to 255, and the grayscale value of each pixel corresponding to the non-root canal filling area may be set to 0.

[0068] In an optional scheme of the present invention, taking into account the grayscale distribution deviation between the oral images to be processed collected by devices of different models, the above-mentioned binarization processing of the oral images to be processed to obtain the binary image also includes: normalizing the grayscale values ​​of the oral images to be processed to obtain the processed oral images; and then binarizing the processed oral images to obtain the binary images.

[0069] Before binarizing the oral cavity image to be processed, the grayscale value of the oral cavity image to be processed may be normalized to improve image contrast and enhance features in the image, thereby making the evaluation result more accurate.

[0070] The grayscale normalization process refers to normalizing the grayscale value of each pixel in the oral image to be processed to 0 to 255. For details, please refer to the following formula:

[0071] Y=(XX min ) / (X max -X min )*255

[0072] Wherein, X represents the original grayscale value of the image, that is, the grayscale value of each pixel in the oral image to be processed, and Y represents the grayscale value after normalization processing.

[0073] After obtaining the binary image, considering that the binary image includes not only the root canal filling area but also the non-root canal filling area, in order to reduce the amount of data processing in subsequent processing and more accurately evaluate the information of the root canal filling area, in an optional solution of the present invention, the following processing can be performed:

[0074] Cropping the binary image to obtain a first image including a root canal filling area;

[0075] Cropping the oral cavity image to be processed to obtain a second image including a root canal filling area;

[0076] According to the oral image to be processed and the binary image, the filling quality of the root canal filling area in the oral image to be processed is evaluated to obtain an evaluation result, including:

[0077] The filling quality of the root canal filling area in the oral cavity image to be processed is evaluated according to the first image and the second image to obtain an evaluation result.

[0078] Among them, the first image corresponding to the root canal filling area in the binary image is cropped for subsequent processing, and the second image corresponding to the root canal filling area in the oral image to be processed is also cropped for subsequent processing. In this way, the first image is a partial image of the binary image, and the second image is a partial image of the oral image to be processed. When the filling quality of the root canal filling area in the oral image to be processed is evaluated based on the first image and the second image, the amount of data processing is reduced. At the same time, since the first image and the second image are both images corresponding to the root canal filling area, they both reflect the characteristics of the root canal filling area, which makes the evaluation result more accurate.

[0079] The cutting can be achieved by any method in the prior art and is not limited in the solution of the present invention.

[0080] In an optional solution of the present invention, the above-mentioned cropping of the binary image to obtain the first image containing the root canal filling area includes:

[0081] According to the binary image, the position of the root canal filling area in the binary image is determined;

[0082] The binary image is cropped according to the position of the root canal filling area in the binary image to obtain a first image including the root canal filling area.

[0083] In an optional scheme of the present invention, since the binary image is first cropped to obtain the first image, and since the second image to be cropped and the first image are both images corresponding to the root canal filling area, in order to ensure that the second image and the first image correspond to the same area, when the oral image to be processed is cropped to obtain the second image including the root canal filling area, it can be specifically done in the following manner: according to the first image, determine the second position of the root canal filling area in the oral image to be processed; according to the second position, crop the oral image to be processed to obtain the second image including the root canal filling area.

[0084] Among them, according to the first image, the second position of the root canal filling area in the oral image to be processed is determined, and the second position refers to the image area corresponding to the root canal filling area in the oral image to be processed. Based on the second position, it can be accurately known which part of the image area corresponds to the root canal filling area in the oral image to be processed, so that the second image obtained by cropping the oral image to be processed based on the second position is the same as the area corresponding to the first image.

[0085] In an optional solution of the present invention, the above-mentioned cropping of the oral cavity image to be processed according to the second position to obtain a second image including the root canal filling area includes:

[0086] According to the second position, generating a cropping bounding box at a position corresponding to the second position in the oral image to be processed;

[0087] Expand the cropping bounding box to obtain an expanded bounding box;

[0088] The oral cavity image to be processed is cropped according to the enlarged bounding box to obtain a second image including the root canal filling area.

[0089] Among them, since the second position refers to the image area corresponding to the root canal filling area in the oral image to be processed, the size of the root canal filling area, that is, the size of the image area corresponding to the root canal filling area, can be known through the second position, and it can also be known which part of the image in the oral image to be processed the root canal filling area is. Then in the scheme of the present invention, a cropping bounding box can be generated according to the second position, the size of the cropping bounding box is the size of the image area corresponding to the root canal filling area, and the position corresponding to the cropping bounding box in the oral image to be processed is the position corresponding to the root canal filling area in the oral image to be processed, then the oral image to be processed can be cropped according to the cropping bounding box to obtain a second image containing the root canal filling area. Considering that there may be some useful information about the filling quality of the root canal filling around the root canal filling area, the cropping bounding box can be expanded, and the oral image to be processed can be cropped by the expanded bounding box, so that the obtained second image includes more information useful for the evaluation result of the filling quality.

[0090] The shape of the cropping bounding box is not limited in the present invention, for example, it can be a regular shape such as a circle, a rectangle, a square, or an irregular shape. The above-mentioned expansion of the cropping bounding box refers to expanding the coverage of the cropping bounding box. The larger the coverage, the larger the size of the obtained second image.

[0091] Optionally, the above-mentioned expansion of the cropping bounding box can be expanded along one boundary direction of the cropping bounding box, or can be expanded along the four sides of the cropping bounding box. For the image corresponding to the cropping bounding box, what is expanded is the number of pixels corresponding to the image. As an example, the second image obtained by cropping the unexpanded cropping bounding box is image a, and the second image obtained by cropping the expanded bounding box is the image corresponding to the pixel points on the four boundaries of image a as the starting point, and 20 pixels are added to the four sides. It should be noted that the above-mentioned increase in the number of pixels refers to increasing the size of the image, not increasing the density of the pixels of the image.

[0092] Step S130 , evaluating the filling quality of the root canal filling area in the oral image to be processed according to the oral image to be processed and the binary image, to obtain an evaluation result.

[0093] Since the oral image to be processed can reflect the group granularity characteristics of the root canal filling area, and the binary image can reflect the fine-grained characteristics of the root canal filling area, the root canal filling information in the oral image to be processed and the root canal filling information in the binary image are comprehensively considered to evaluate the quality of the root canal filling, which can improve the quality assessment results.

[0094] The evaluation result may be just-filled, under-filled or over-filled, and the evaluation result may provide the doctor with data basis for the next step of processing the root canal filling area.

[0095] Among them, just filling means that the root canal filling material just fills the pulp cavity, underfilling means that the root canal filling material does not fill the pulp cavity, and overfilling means that the root canal filling material overflows the pulp cavity.

[0096] In an optional solution of the present invention, the filling quality of the root canal filling area in the oral image to be processed is evaluated based on the oral image to be processed and the binary image to obtain the evaluation result, including:

[0097] Extracting a first image feature of the oral image to be processed and a second image feature of the binary image respectively, wherein the first image feature is a regional feature of the root canal filling area, and the second image feature is an edge feature of the root canal filling area;

[0098] The filling quality of the root canal filling area in the oral cavity image to be processed is evaluated according to the first image feature and the second image feature to obtain an evaluation result.

[0099] The regional feature refers to the feature reflecting the entire root canal filling area, and the edge feature refers to the feature reflecting the edge of the root canal filling area. Since the root canal filling quality is affected by the filling gap between the root canal edge and the pulp cavity, in the solution of the present invention, the regional feature and the edge feature of the root canal filling area are combined, so that when the filling quality of the root canal filling area in the oral cavity image to be processed is evaluated based on the first image feature and the second image feature, the evaluation result is more accurate.

[0100] It should be noted that, in the solution of the present invention, quality assessment may also be performed based on only one of the first image feature or the second image feature, which is within the protection scope of the present invention.

[0101] In an optional scheme of the present invention, the filling quality of the root canal filling area in the oral image to be processed is evaluated according to the first image feature and the second image feature, and an implementation method for obtaining the evaluation result is: according to the first image feature and the second image feature, the distance between the root canal filling area and the pulp cavity wall is calculated, and the distance is used to characterize whether there is a filling gap between the pulp cavity walls of the root canal filling area; when the distance is less than a set threshold and the corresponding root canal filling material in the root canal filling area is in the pulp cavity, it indicates that there is no filling gap, and the corresponding evaluation result is just filling; when the distance is not less than the set threshold, it indicates that there is a filling gap, and the corresponding evaluation result is under-filling; when the distance is less than the set threshold and the corresponding root canal filling material in the root canal filling area overflows the pulp cavity, the corresponding filling result is over-filling.

[0102] In an optional embodiment of the present invention, the above-mentioned processing of cropping the binary image to obtain the first image containing the root canal filling area is implemented by a root canal filling segmentation model, wherein the above-mentioned root canal filling segmentation model is trained in the following manner:

[0103] Acquire multiple sample images, each sample image including a periapical film after the root canal is filled;

[0104] Perform binary processing on each sample image to obtain a binary image, wherein each pixel point in each binary image corresponds to a segmentation and annotation result, and for each pixel point in each binary image, the segmentation and annotation result represents whether the pixel point belongs to a root canal filling area or a non-root canal filling area in the binary image;

[0105] The first initial model is trained according to each binary image until the loss value of the first initial model meets the first training end condition, and the first initial model corresponding to the first training end condition is used as the root canal filling segmentation model.

[0106] Among them, the input of the first initial model is each binary image, and the output is the predicted segmentation result corresponding to each binary image. The loss value of the first initial model represents the difference between the predicted segmentation result corresponding to each binary image and the segmentation annotation result.

[0107] Among them, the binary image can be cropped by the pre-trained root canal filling segmentation model to obtain the first image. The root canal filling segmentation model trained by the first initial model of the neural network can accurately identify the position of the root canal filling area in the binary image when cropping the binary image, so that the first image can be accurately cropped from the binary image.

[0108] Optionally, before performing binary processing on each sample image to obtain a binary image, grayscale value normalization processing may be performed on each sample image to obtain a processed sample image.

[0109] Among them, the binary image can be labeled by a pixel-by-pixel labeling method for each processed sample image to obtain a segmentation labeling result, that is, each pixel in each binary image corresponds to a segmentation labeling result, and for each pixel in each binary image, the segmentation labeling result characterizes whether the pixel belongs to the root canal filling area or the non-root canal filling area in the binary image. Through the pixel-by-pixel labeling method, it is possible to distinguish which pixels in the binary image belong to the root canal filling area and which pixels belong to the non-root canal filling area at the pixel level, so that when the model obtained by training the binary image after pixel-by-pixel labeling is applied, it can accurately distinguish which part of the image is the root canal filling area and which part is the non-root canal filling area at the pixel level. Among them, each processed sample image after labeling is a binary image. For the convenience of description, the labeled binary image is referred to as a binary sample image below.

[0110] In the solution of the present invention, each binary sample image can be divided into a training set, a validation set and a test set. As an example, the ratio of the number of sample images in the training set, the validation set and the test set is 8:1:1.

[0111] The initial neural network model (first initial model) is trained according to each binary sample image in the training set until the loss value of the initial neural network model meets the first training end condition, and the neural network model corresponding to the first training end condition is used as the root canal filling segmentation model. The output of the above initial neural network model is the predicted segmentation result of each binary sample image. For each binary sample image, the predicted segmentation result corresponding to the binary sample image can be the position of the root canal filling area in the binary sample image. The above loss value represents the difference between the predicted segmentation result of each binary sample image and the corresponding annotation result. Among them, the first training end condition can be configured based on actual needs, for example, the loss value is less than the first set value, or within the first range.

[0112] During the training process, the parameters of the initial neural network model can be continuously optimized through the validation set and the test set, so that the root canal filling segmentation model finally trained has higher accuracy.

[0113] As an example, the number of training iterations can be set to 50, the initial learning rate of the initial neural network model is 0.0003, the confidence threshold corresponding to the root canal filling area and the non-root canal filling area is 0.5, and after testing the root canal filling segmentation model with 150 test samples, the obtained similarity coefficient Dice (Dice Similarity Coefficient) value is 0.924, where the Dice coefficient is a set similarity measurement indicator, which is usually used to calculate the similarity of two samples, with a value range of 0-1. The best segmentation result is 1, and the worst segmentation result is 0. The Dice value after the test is 0.924, indicating that the trained root canal filling segmentation model has good accuracy.

[0114] Optionally, the loss function corresponding to the above initial neural network model may be a combined loss function of Cross-entropy loss and Dice Loss.

[0115] Optionally, before training the initial neural network model according to the binary sample images in the training set, each binary sample image may be resized to facilitate processing of the initial neural network model. As an example, the size of the binary sample image may be normalized to 512*512.

[0116] Optionally, the above training process may use an optimizer to speed up the training of the model, wherein the optimizer may be any one of Adam (Adaptive moment estimation), SGD (Stochastic Gradient Descent) optimizer or AdamW (Adam Weight Decay Optimizer) optimizer.

[0117] Optionally, the above-mentioned initial neural network model can be any one of RCFS-Net, FCN (Fully Convolutional Networks for Semantic Segmentation), DeepLab, U-Net++, YOLO (You Only Look Once), SSD (Single Shot MultiBox Detector), Fast R-CNN (Fast Region-Convolutional Neural Network), and Faster R-CNN (Faster Region-Convolutional Neural Network), which is not limited in the solution of the present invention.

[0118] In an optional solution of the present invention, the process of evaluating the filling quality of the root canal filling area in the oral image to be processed according to the first image and the second image to obtain the evaluation result is implemented by a root canal filling quality category recognition model, wherein the root canal filling quality category recognition model is trained in the following manner:

[0119] Acquire multiple sample images, each sample image corresponds to a quality annotation result, and the quality annotation result is any one of just filling, underfilling or overfilling;

[0120] For each sample image, determine a first image and a second image corresponding to the sample image, and regard the first image and the second image corresponding to each sample image as a sample image pair;

[0121] According to each sample image pair, the second initial model is trained until the loss value of the second initial model meets the second training end condition, and the second initial model corresponding to the second training end condition is used as the root canal filling quality category recognition model;

[0122] Among them, the input of the second initial model is each sample image pair, and the output is the prediction quality assessment result corresponding to each sample image pair. The loss value of the second initial model represents the difference between the prediction quality assessment result corresponding to each sample image pair and the quality labeling result.

[0123] Among them, the root canal filling quality category recognition model obtained by pre-training can be used to implement quality assessment of the filling quality of the root canal filling area in the processed oral image to obtain an evaluation result. The root canal filling quality category recognition model obtained by neural network training can accurately extract the features corresponding to the root canal filling area in the first image and the second image when performing quality assessment on the filling quality of the root canal filling area in the processed oral image, thereby accurately performing quality assessment.

[0124] Optionally, during training, each sample image pair may be divided into a training set, a validation set, and a test set. As an example, the ratio of the number of sample image pairs in the training set, the validation set, and the test set is 8:1:1.

[0125] The initial network model (second initial model) is trained according to each sample image pair in the training set until the loss value of the initial network model meets the second training end condition, and the network model corresponding to the second training end condition is used as the root canal filling quality category recognition model. The output of the above initial network model is the predicted classification result of each sample image pair. For each pair of sample image pairs, the predicted classification result corresponding to the sample image pair can be represented by probability, and the above loss value represents the difference between the predicted classification result of each sample image pair and the corresponding quality labeling result. Among them, the second training end condition can be configured based on actual needs, for example, the loss value is less than the second set value, or within the second range. The above first setting value and the second setting value may be the same or different. Similarly, the first range and the second range may be the same or different.

[0126] During the training process, the parameters of the initial network model can be continuously optimized through the validation set and the test set, so that the root canal filling quality category recognition model finally trained has higher accuracy.

[0127] The root canal filling quality category recognition model is a three-category model, corresponding to the three classification results of just filling, underfilling and overfilling respectively.

[0128] As an example, the number of training iterations in the training process of the above-mentioned root canal filling quality category recognition model can be set to 50, the initial learning rate of the initial network model is 0.0001, and after testing the root canal filling quality category recognition model with 150 test samples, the obtained similarity coefficient Dice value is 0.947, indicating that the trained root canal filling quality category recognition model has good accuracy.

[0129] Optionally, before training the initial network model according to the sample images in the training set, the size of each image in each sample image pair can be normalized to facilitate the processing of the initial network model. As an example, the size of each image in each sample image pair can be normalized to 448*448.

[0130] Optionally, the above training process may use an optimizer to speed up the training of the model, wherein the optimizer may be any one of Adam (Adaptive moment estimation), SGD (Stochastic Gradient Descent) optimizer or AdamW (Adam Weight Decay Optimizer) optimizer.

[0131] Optionally, the above-mentioned initial network model can be any one of RCFC-Net, VGG (Visual Geometry Group Network), ResNet (Residual Network), ResNeXt, DenseNet (DENSELYCONNECTED NETWORKS), EfficientNet, and RegNet, which is not limited in the solution of the present invention.

[0132] It should be noted that the root canal filling quality category identification model and the root canal filling segmentation model may be one model or two models, which is not limited in the solution of the present invention.

[0133] In order to better illustrate and understand the principle of the method provided by the present invention, the scheme of the present invention is described below in conjunction with an optional specific embodiment. It should be noted that the specific implementation of each step in the specific embodiment should not be understood as a limitation to the scheme of the present invention. On the basis of the principle of the scheme provided by the present invention, other implementations that can be thought of by those skilled in the art should also be considered within the scope of protection of the present invention.

[0134] In this example, according to the model training method described above, two models are pre-trained, namely the root canal filling quality category recognition model and the above-mentioned root canal filling segmentation model.

[0135] See also Figure 2 The flow chart of the oral image processing method shown in FIG. 1 is a flow chart of the oral image processing method of the present invention, and the oral image processing method of the present invention comprises the following steps:

[0136] Step 1: Obtain the oral cavity image to be processed ( Figure 2 Oral periapical radiographs in the figure).

[0137] Step 2: normalize the grayscale value of the oral image to be processed to obtain the processed oral image (corresponding to Figure 2 The grayscale normalization step of the image).

[0138] Step 3, binarizing the processed oral image to obtain a binary image, and inputting the binary image into the root canal filling segmentation model to obtain a first image containing the root canal filling area (corresponding to Figure 2 Steps of training the root canal filling segmentation model RCFS-Net).

[0139] Step 4: Generate a cropping bounding box at a position corresponding to the second position in the processed oral cavity image based on the first image.

[0140] Step 5: Expand the cropped bounding box outwards to a length of 20 pixels to obtain an enlarged bounding box.

[0141] Step 6: based on the enlarged bounding box, the processed oral image is cropped to obtain a second image (step 5 and step 6 correspond to Figure 2 Steps for cropping the root canal filling area).

[0142] Step 7: Input the first image and the second image into the root canal filling quality category recognition model to obtain an evaluation result (corresponding to Figure 2 The trained root canal filling quality assessment RCFC-Net and the steps of outputting the root canal filling quality category).

[0143] According to the solution of the present invention, two second-order neural network models (root canal filling quality category recognition model and root canal filling segmentation model) are used, that is, the input of the root canal filling quality category recognition model is determined according to the output of the root canal filling segmentation model, and the evaluation result is obtained through the two models, so that the accuracy of the evaluation result can be improved. Moreover, in the solution of the present invention, the input of the root canal filling quality category recognition model is a two-channel input, one channel is the first image, and the other channel is the second image, and the evaluation result obtained through these two images is more accurate.

[0144] Based on Figure 1 Based on the same principle as the method shown in , the embodiment of the present invention also provides an oral image processing device 20, such as Figure 3 As shown in , the oral image processing device 20 may include an image acquisition module 210, a binarization module 220 and a quality assessment module 230, wherein:

[0145] An image acquisition module 210 is used to acquire an oral cavity image to be processed, wherein the oral cavity image to be processed includes a periapical film after root canal filling;

[0146] A binarization module 220, used for performing binarization processing on the oral cavity image to be processed to obtain a binary image;

[0147] The quality assessment module 230 is used to assess the filling quality of the root canal filling area in the oral image to be processed based on the oral image to be processed and the binary image to obtain an assessment result.

[0148] Optionally, the device further comprises:

[0149] A cropping module is used to crop the binary image to obtain a first image including the root canal filling area; and to crop the oral cavity image to be processed to obtain a second image including the root canal filling area;

[0150] The quality assessment module 230 is specifically used to:

[0151] The filling quality of the root canal filling area in the oral cavity image to be processed is evaluated according to the first image and the second image to obtain an evaluation result.

[0152] Optionally, when the above-mentioned cropping module crops the binary image to obtain the first image including the root canal filling area, it is specifically used to:

[0153] According to the binary image, the position of the root canal filling area in the binary image is determined;

[0154] According to the position of the root canal filling area in the binary image, the binary image is cropped to obtain a first image including the root canal filling area;

[0155] When the above-mentioned cropping module crops the oral cavity image to be processed to obtain the second image including the root canal filling area, it is specifically used to:

[0156] Determining a second position of the root canal filling area in the oral cavity image to be processed according to the first image;

[0157] The oral cavity image to be processed is cropped according to the second position to obtain a second image including the root canal filling area.

[0158] Optionally, when the cropping module crops the oral cavity image to be processed according to the second position to obtain a second image including the root canal filling area, it is specifically used to:

[0159] According to the second position, generating a cropping bounding box at a position corresponding to the second position in the oral image to be processed;

[0160] Expand the cropping bounding box to obtain an expanded bounding box;

[0161] The oral cavity image to be processed is cropped according to the enlarged bounding box to obtain a second image including the root canal filling area.

[0162] Optionally, when the above-mentioned cropping module evaluates the filling quality of the root canal filling area in the oral image to be processed according to the oral image to be processed and the binary image to obtain the evaluation result, it is specifically used to:

[0163] Extracting a first image feature of the oral image to be processed and a second image feature of the binary image respectively, wherein the first image feature is a regional feature of the root canal filling area, and the second image feature is an edge feature of the root canal filling area;

[0164] The filling quality of the root canal filling area in the oral cavity image to be processed is evaluated according to the first image feature and the second image feature to obtain an evaluation result.

[0165] Optionally, the above-mentioned processing of cropping the binary image to obtain the first image containing the root canal filling area is implemented by a root canal filling segmentation model, wherein the above-mentioned root canal filling segmentation model is trained in the following manner:

[0166] Acquire multiple sample images, each sample image including a periapical film after the root canal is filled;

[0167] Binary processing is performed on each sample image to obtain a binary image, each pixel point in each binary image corresponds to a segmentation and annotation result, and for each pixel point in each binary image, the segmentation and annotation result represents whether the pixel point belongs to a root canal filling area or a non-root canal filling area in the binary image;

[0168] The first initial model is trained according to each binary image until the loss value of the first initial model meets the first training end condition, and the first initial model corresponding to the first training end condition is used as the root canal filling segmentation model.

[0169] The input of the first initial model is each binary image, and the output is the predicted segmentation result corresponding to each binary image. The loss value of the first initial model represents the difference between the predicted segmentation result corresponding to each binary image and the segmentation annotation result.

[0170] The above-mentioned process of evaluating the filling quality of the root canal filling area in the oral image to be processed according to the first image and the second image to obtain the evaluation result is realized by a root canal filling quality category recognition model, wherein the above-mentioned root canal filling quality category recognition model is trained in the following manner:

[0171] Acquire multiple sample images, each sample image corresponds to a quality annotation result, and the quality annotation result is any one of just filling, underfilling or overfilling;

[0172] For each sample image, determine a first image and a second image corresponding to the sample image, and regard the first image and the second image corresponding to each sample image as a sample image pair;

[0173] According to each sample image pair, the second initial model is trained until the loss value of the second initial model meets the second training end condition, and the second initial model corresponding to the second training end condition is used as the root canal filling quality category recognition model;

[0174] Among them, the input of the second initial model is each sample image pair, and the output is the prediction quality assessment result corresponding to each sample image pair. The loss value of the second initial model represents the difference between the prediction quality assessment result corresponding to each sample image pair and the quality labeling result.

[0175] Optionally, before the binarization module 220 performs binarization processing on the oral image to be processed to obtain a binary image, the device further includes:

[0176] The preprocessing module performs grayscale normalization processing on the oral image to be processed to obtain a processed oral image.

[0177] The oral image processing device of the embodiment of the present invention can execute the oral image processing method provided by the embodiment of the present invention, and the implementation principle is similar. The actions performed by each module and unit in the oral image processing device in each embodiment of the present invention correspond to the steps in the oral image processing method in each embodiment of the present invention. For the detailed functional description of each module of the oral image processing device, please refer to the description in the corresponding oral image processing method shown in the previous text, which will not be repeated here.

[0178] Among them, the above-mentioned oral image processing device can be a computer program (including program code) running in a computer device, for example, the oral image processing device is an application software; the device can be used to execute the corresponding steps in the method provided in the embodiment of the present invention.

[0179] In some embodiments, the oral image processing device provided by the embodiment of the present invention can be implemented by a combination of software and hardware. As an example, the oral image processing device provided by the embodiment of the present invention can be a processor in the form of a hardware decoding processor, which is programmed to execute the oral image processing method provided by the embodiment of the present invention. For example, the processor in the form of a hardware decoding processor can adopt one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field programmable gate arrays (FPGAs) or other electronic components.

[0180] In other embodiments, the oral image processing device provided by the embodiments of the present invention may be implemented in a software manner. Figure 3 An oral image processing device stored in a memory is shown, which may be software in the form of a program and a plug-in, and includes a series of modules, including an image acquisition module 210, a binarization module 220 and a quality assessment module 230, for implementing the oral image processing method provided in an embodiment of the present invention.

[0181] The modules involved in the embodiments of the present invention may be implemented by software or hardware, wherein the name of a module does not limit the module itself in some cases.

[0182] Based on the same principle as the method shown in the embodiments of the present invention, an electronic device is also provided in the embodiments of the present invention, which may include but is not limited to: a processor and a memory; the memory is used to store a computer program; the processor is used to execute the method shown in any embodiment of the present invention by calling the computer program.

[0183] In an alternative embodiment, an electronic device is provided, such as Figure 4 As shown, Figure 4 The electronic device 4000 shown includes: a processor 4001 and a memory 4003. The processor 4001 and the memory 4003 are connected, such as through a bus 4002. Optionally, the electronic device 4000 may also include a transceiver 4004, which may be used for data interaction between the electronic device and other electronic devices, such as data transmission and / or data reception. It should be noted that in actual applications, the transceiver 4004 is not limited to one, and the structure of the electronic device 4000 does not constitute a limitation on the embodiments of the present invention.

[0184] Processor 4001 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. It may implement or execute various exemplary logic blocks, modules and circuits described in conjunction with the disclosure of the present invention. Processor 4001 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0185] The bus 4002 may include a path to transmit information between the above components. The bus 4002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus 4002 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0186] The memory 4003 can be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compressed optical disk, laser disk, optical disk, digital versatile disk, Blu-ray disk, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited to this.

[0187] The memory 4003 is used to store the application code (computer program) for executing the solution of the present invention, and the execution is controlled by the processor 4001. The processor 4001 is used to execute the application code stored in the memory 4003 to implement the content shown in the above method embodiment.

[0188] The electronic device may also be a terminal device. Figure 4 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0189] An embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored. When the computer-readable storage medium is run on a computer, the computer can execute the corresponding contents of the aforementioned method embodiment.

[0190] According to another aspect of the present invention, a computer program product or a computer program is also provided, the computer program product or the computer program comprising computer instructions, the computer instructions being stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the methods provided in the implementation modes of the above-mentioned various embodiments.

[0191] Computer program code for performing the operations of the present invention may be written in one or more programming languages, or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0192] It should be understood that the flow charts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the module, the program segment, or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a sequence different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0193] The computer-readable storage medium provided by the embodiment of the present invention may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to, an electrical connection with one or more wires, 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 thereof. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, device, or device.

[0194] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device executes the method shown in the above embodiment.

[0195] The above description is only a preferred embodiment of the present invention and an explanation of the technical principles used. Those skilled in the art should understand that the disclosure scope involved in the present invention is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present invention (but not limited to) by each other.

Claims

1. An oral image processing method, characterized in that: include: Acquire an oral cavity image to be processed, wherein the oral cavity image to be processed includes a periapical film after root canal filling; Binarizing the oral cavity image to be processed to obtain a binary image; According to the oral image to be processed and the binary image, evaluating the filling quality of the root canal filling area in the oral image to be processed to obtain an evaluation result; Also includes: Cropping the binary image to obtain a first image including a root canal filling area; Cropping the oral cavity image to be processed to obtain a second image including the root canal filling area; The step of evaluating the filling quality of the root canal filling area in the oral image to be processed according to the oral image to be processed and the binary image to obtain an evaluation result includes: Extracting a first image feature of the oral image to be processed and a second image feature of the binary image respectively, wherein the first image feature is a regional feature of the root canal filling area, and the second image feature is an edge feature of the root canal filling area; The filling quality of the root canal filling area in the oral cavity image to be processed is evaluated according to the first image feature and the second image feature to obtain an evaluation result.

2. The method according to claim 1, characterized in that The step of cropping the binary image to obtain a first image including a root canal filling area includes: Determining a first position of the root canal filling area in the binary image according to the binary image; According to the first position, the binary image is cropped to obtain a first image including a root canal filling area; The step of cropping the oral cavity image to be processed to obtain a second image including the root canal filling area includes: Determining a second position of the root canal filling area in the oral cavity image to be processed according to the first image; The oral cavity image to be processed is cropped according to the second position to obtain a second image including the root canal filling area.

3. The method according to claim 2, characterized in that The method of cropping the oral cavity image to be processed according to the second position to obtain a second image including the root canal filling area includes: According to the second position, generating a cropping bounding box at a position corresponding to the second position in the oral image to be processed; Expanding the cropping bounding box to obtain an expanded bounding box; The oral cavity image to be processed is cropped according to the enlarged bounding box to obtain a second image including the root canal filling area.

4. The method according to any one of claims 1 to 3, characterized in that The processing of cropping the binary image to obtain the first image containing the root canal filling area is implemented by a root canal filling segmentation model, wherein the root canal filling segmentation model is trained in the following manner: Acquire multiple sample images, each sample image including a periapical film after the root canal is filled; Performing binary processing on each of the sample images to obtain a binary image, wherein each pixel point in each of the binary images corresponds to a segmentation and annotation result, and for each pixel point in each of the binary images, the segmentation and annotation result indicates whether the pixel point belongs to a root canal filling area or a non-root canal filling area in the binary image; The first initial model is trained according to each of the binary images until the loss value of the first initial model satisfies a first training end condition, and the first initial model corresponding to the first training end condition is used as a root canal filling segmentation model. The input of the first initial model is each binary image, and the output is the predicted segmentation result corresponding to each binary image, and the loss value of the first initial model represents the difference between the predicted segmentation result corresponding to each binary image and the segmentation annotation result; The process of evaluating the filling quality of the root canal filling area in the oral image to be processed according to the first image and the second image to obtain the evaluation result is implemented by a root canal filling quality category recognition model, wherein the root canal filling quality category recognition model is trained in the following manner: Acquire multiple sample images, each of the sample images corresponds to a quality annotation result, and the quality annotation result is any one of just filling, underfilling or overfilling; For each of the sample images, determining a first image and a second image corresponding to the sample image, and treating the first image and the second image corresponding to each of the sample images as a sample image pair; According to each of the sample image pairs, the second initial model is trained until the loss value of the second initial model satisfies a second training end condition, and the second initial model corresponding to the second training end condition is used as the root canal filling quality category recognition model; Among them, the input of the second initial model is each sample image pair, and the output is the prediction quality assessment result corresponding to each sample image pair. The loss value of the second initial model represents the difference between the prediction quality assessment result corresponding to each sample image pair and the quality labeling result.

5. The method according to any one of claims 1 to 3, characterized in that Before the to-be-processed oral cavity image is binarized to obtain a binary image, the method further includes: Grayscale normalization is performed on the oral cavity image to be processed to obtain a processed oral cavity image.

6. An oral image processing device, characterized in that: The oral image processing method according to claim 1 is adopted, and the device comprises: An image acquisition module, used for acquiring an oral cavity image to be processed, wherein the oral cavity image to be processed includes a periapical film after root canal filling; A binarization module, used for performing binarization processing on the oral cavity image to be processed to obtain a binary image; The quality assessment module is used to assess the filling quality of the root canal filling area in the oral image to be processed based on the oral image to be processed and the binary image to obtain an assessment result.

7. An electronic device, characterized in that: The method comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method according to any one of claims 1 to 5 when executing the computer program.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images