A method, device, equipment and storage medium for processing panoramic dental images

Abstract

The present disclosure provides a method, apparatus, device and storage medium for processing panoramic dental images. The method mainly includes: obtaining an image to be processed, where the image to be processed is a panoramic dental image; identifying the teeth in the image to be processed according to a first recognition model to obtain a first recognition result, where the first recognition result includes the first position information and binary masks of all the teeth in the image to be processed; numbering all the teeth in the image to be processed according to the first recognition result to obtain tooth numbering information, where the tooth numbering information includes tooth numbers and the first position information of the teeth corresponding to the tooth numbers; identifying abnormal teeth in the image to be processed according to a second recognition model to obtain a second recognition result, where the second recognition result includes the second position information and image features of the abnormal teeth; and generating a processing result of the panoramic dental image according to the tooth numbering information and the second recognition result.

Description

Technical Field

[0001] The present disclosure relates to the field of computer vision technology, and in particular, to a method, device, equipment, and storage medium for processing panoramic dental images. Background Art

[0002] Panoramic dental images can fully reflect the tooth morphology. By analyzing information such as the degree of tooth damage and color change in panoramic dental images, it is possible to help judge the dental health status. However, the above operations in related technologies mainly rely on manual labor, that is, doctors observe panoramic dental images with the naked eye and analyze and judge the dental health status based on experience. Therefore, related technologies have problems of wasting manpower and low efficiency, and the analysis and judgment results strongly depend on the doctor's knowledge and experience, which also leads to low accuracy of the results relying on manual analysis and judgment in related technologies. Summary of the Invention

[0003] The present disclosure provides a method, device, equipment, and storage medium for processing panoramic dental images to at least solve the above technical problems existing in the prior art.

[0004] According to a first aspect of the present disclosure, a method for processing panoramic dental images is provided. The method includes: obtaining an image to be processed, where the image to be processed is a panoramic dental image; identifying teeth in the image to be processed according to a first recognition model to obtain a first recognition result, where the first recognition result includes first position information and a binary mask of all teeth in the image to be processed; numbering all teeth in the image to be processed according to the first recognition result to obtain tooth numbering information, where the tooth numbering information includes tooth numbers and first position information of teeth corresponding to the tooth numbers; identifying abnormal teeth in the image to be processed according to a second recognition model to obtain a second recognition result, where the second recognition result includes second position information and image features of the abnormal teeth; and generating a processing result of the panoramic dental image according to the tooth numbering information and the second recognition result.

[0005] In an implementable embodiment, before identifying teeth in the image to be processed according to the first recognition model to obtain a first recognition result, the method further includes: obtaining a first sample data set, where the first sample data set includes panoramic dental sample images; performing tooth annotation on the panoramic dental sample images in the first sample data set to obtain first training samples; and training a convolutional neural network according to the first training samples to obtain the first recognition model.

[0006] In one possible implementation, according to the first recognition result, all teeth in the image to be processed are numbered to obtain tooth numbering information, including: grouping all teeth in the image to be processed according to the first position information to obtain a grouping result; numbering all teeth in the image to be processed according to the grouping result to obtain the tooth numbering information.

[0007] In one possible implementation, the grouping all teeth in the image to be processed according to the first position information to obtain a grouping result includes: sorting all teeth in the image to be processed in ascending order according to the abscissa in the first position information to obtain a sorting result; sequentially calculating the vertical coordinate centers of a specified number of teeth according to the sorting result; calculating the average value of the vertical coordinate centers of the specified number of teeth; and grouping all teeth in the image to be processed according to the vertical coordinate in the first position information and the average value to obtain a grouping result.

[0008] In one possible implementation, the grouping all teeth in the image to be processed according to the vertical coordinate in the first position information and the average value to obtain a grouping result includes: calculating the difference between the vertical coordinate in the first position information and the average value to obtain a first calculation result; if the first calculation result is negative, classifying the teeth corresponding to the vertical coordinate in the first position information into the upper tooth group; if the first calculation result is positive, classifying the teeth corresponding to the vertical coordinate in the first position information into the lower tooth group; and grouping all teeth in the image to be processed according to the upper tooth group and the lower tooth group to obtain a grouping result.

[0009] In one possible implementation, after numbering all teeth in the image to be processed according to the first recognition result to obtain tooth numbering information, the method further includes: parsing the binary mask of all teeth in the image to be processed to obtain a tooth binary image; performing an AND operation on the tooth binary image and the image to be processed to obtain the original image of all teeth in the image to be processed; performing background transparency processing on the original image to obtain an original image with a transparent background; and saving the original image with the transparent background under the name corresponding to its tooth number.

[0010] In one possible implementation, recognizing abnormal teeth in the image to be processed according to the second recognition model to obtain a second recognition result includes: extracting features of the image to be processed according to the second recognition model to obtain a feature map of the image to be processed; locating abnormal teeth in the image to be processed according to the feature map of the image to be processed to obtain second position information of the abnormal teeth; and outputting the second position information of the abnormal teeth and the image features of the abnormal teeth corresponding to the second recognition model.

[0011] In one implementable manner, generating a dental panoramic image processing result according to the tooth labeling information and the second recognition result includes: sequentially calculating the distances between the second position information of the abnormal tooth and the first position information of all teeth in the image to be processed, obtaining a second calculation result; selecting the minimum distance between the second position information of the abnormal tooth and the first position information of all teeth in the image to be processed according to the second calculation result; binding the second recognition result with the tooth labeling information corresponding to the minimum distance to generate the dental panoramic image processing result.

[0012] According to a second aspect of the present disclosure, there is provided a dental panoramic image processing apparatus, characterized in that the apparatus includes: an acquisition module for acquiring an image to be processed, where the image to be processed is a dental panoramic image; a first recognition module for recognizing teeth in the image to be processed according to a first recognition model to obtain a first recognition result, where the first recognition result includes the first position information and a binary mask of all teeth in the image to be processed; a labeling module for labeling all teeth in the image to be processed according to the first recognition result to obtain tooth labeling information, where the tooth labeling information includes tooth labels and the first position information of the teeth corresponding to the tooth labels; a second recognition module for recognizing abnormal teeth in the image to be processed according to a second recognition model to obtain a second recognition result, where the second recognition result includes the second position information and image features of the abnormal teeth; and a generation module for generating a dental panoramic image processing result according to the tooth labeling information and the second recognition result.

[0013] According to a third aspect of the present disclosure, there is provided an electronic device, including:

[0014] at least one processor; and

[0015] a memory communicatively connected to the at least one processor; wherein,

[0016] the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method of the present disclosure.

[0017] According to a fourth aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium storing computer instructions for causing a computer to execute the method of the present disclosure.

[0018] A method, device, equipment and storage medium for processing panoramic dental images according to the present disclosure can identify abnormal teeth in panoramic dental images through model recognition and output the labels and image features of the abnormal teeth, thereby maximizing the extraction of useful information from panoramic dental images for doctors to judge the dental health status, further improving the judgment efficiency and the accuracy of judgment results.

[0019] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it used to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] By referring to the drawings and reading the following detailed description, the above and other objects, features and advantages of the exemplary embodiments of the present disclosure will become easily understood. In the drawings, several embodiments of the present disclosure are shown in an exemplary rather than restrictive manner, where:

[0021] In the drawings, the same or corresponding reference numerals represent the same or corresponding parts.

[0022] Figure 1 FIG. shows a schematic flowchart of a method for processing panoramic dental images according to the first embodiment of the present disclosure;

[0023] Figure 2 FIG. shows a schematic diagram of tooth labels of a method for processing panoramic dental images according to the first embodiment of the present disclosure;

[0024] Figure 3 FIG. shows a schematic flowchart of a method for processing panoramic dental images according to the second embodiment of the present disclosure;

[0025] Figure 4 FIG. shows a schematic flowchart of a method for processing panoramic dental images according to the third embodiment of the present disclosure;

[0026] Figure 5 FIG. shows a schematic flowchart of a method for processing panoramic dental images according to the fourth embodiment of the present disclosure;

[0027] Figure 6 FIG. shows a schematic diagram of position information of a method for processing panoramic dental images according to the fourth embodiment of the present disclosure;

[0028] Figure 7 FIG. shows a schematic flowchart of a method for processing panoramic dental images according to the fifth embodiment of the present disclosure;

[0029] Figure 8 FIG. shows a schematic flowchart of a method for processing panoramic dental images according to the sixth embodiment of the present disclosure;

[0030] Figure 9 The figure shows a schematic flowchart of a method for processing panoramic dental images according to the seventh embodiment of the present disclosure;

[0031] Figure 10 The figure shows a schematic flowchart of a method for processing panoramic dental images according to the eighth embodiment of the present disclosure;

[0032] Figure 11 The figure shows a schematic structural diagram of a panoramic dental image processing device according to the ninth embodiment of the present disclosure;

[0033] Figure 12 The figure shows a schematic composition structure diagram of an electronic device according to an embodiment of the present disclosure. Detailed implementation manners

[0034] To make the objectives, features, and advantages of the present disclosure more obvious and understandable, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, rather than all the embodiments. Based on the embodiments of the present disclosure, all other embodiments obtained by those skilled in the art without creative efforts shall fall within the protection scope of the present disclosure.

[0035] Figure 1 The figure is a schematic flowchart of a method for processing panoramic dental images according to the first embodiment of the present disclosure. As Figure 1 shown, the method mainly includes:

[0036] Step S101: Obtain an image to be processed, where the image to be processed is a panoramic dental image.

[0037] In this embodiment, first, a panoramic dental image should be obtained as the image to be processed. The panoramic dental image is an image of all teeth and surrounding tissues obtained by applying the principle of narrow-slit and circular-arc orbit tomography and through one-time imaging. The panoramic dental image can fully reflect the shape of the teeth. By analyzing information such as the degree of tooth damage and color change in the panoramic dental image, the dental health status can be judged.

[0038] Step S102: Identify the teeth in the image to be processed according to the first recognition model to obtain a first recognition result, where the first recognition result includes the first position information and binary mask of all teeth in the image to be processed.

[0039] In this embodiment, after obtaining the image to be processed, it is necessary to identify and segment single teeth in the image to be processed according to the first recognition model, so as to obtain the first position information and binary mask of all teeth in the image to be processed. The first position information can be expressed as [x, y, width, height]; where x is the abscissa of the upper left vertex of the tooth detection frame of a single tooth, y is the ordinate of the upper left vertex of the tooth detection frame of a single tooth, width is the width of the tooth detection frame of a single tooth, and height is the height of the tooth detection frame of a single tooth. The binary map data after decoding the binary mask can completely depict the single tooth in the image to be processed, eliminate the interference of information of non-tooth parts, and enhance the readability of the information.

[0040] In an implementable manner, the first recognition model can be a convolutional neural network model. The first recognition model first performs convolution on the image to be processed to extract the features of the image to be processed, and then, according to the features of the image to be processed, identifies and segments single teeth in the image to be processed, and returns the first position information and binary mask of each tooth.

[0041] Step S103: Number all the teeth in the image to be processed according to the first recognition result to obtain tooth numbering information, where the tooth numbering information includes the tooth number and the first position information of the tooth corresponding to the tooth number.

[0042] In this embodiment, in order to facilitate the distinction of single teeth in the image to be processed, it is necessary to number all the teeth in the image to be processed. Specifically, all the teeth in the image to be processed can be numbered according to the first recognition result to obtain tooth numbering information, where the tooth numbering information includes the tooth number and the first position information of the tooth corresponding to the tooth number, and the tooth numbering information can be saved in the form of a data dictionary, such as [tooth number: first position information].

[0043] In an implementable manner, the teeth in the image to be processed can be numbered according to the Fédération Dentaire Internationale (abbreviated as FDI) tooth position representation method. The FDI tooth position representation method divides teeth into four groups, namely upper left (Quadrant I), upper right (Quadrant II), lower right (Quadrant III), and lower left (Quadrant IV). Each tooth is represented by two Arabic numerals. The first digit represents the quadrant where the tooth is located. If the tooth is a permanent tooth, the upper left, upper right, lower right, and lower left are 1, 2, 3, and 4 respectively. If the tooth is a deciduous tooth, the upper left, upper right, lower right, and lower left are 5, 6, 7, and 8 respectively. The second digit represents the position of the tooth, which is 1-8 from the central incisor to the third molar.

[0044] Figure 2 It is a schematic diagram of tooth numbering for a method for processing panoramic dental images according to the first embodiment of the present disclosure, as Figure 2As shown, if the tooth is a permanent tooth, the tooth numbers in the first quadrant are 18, 17, 16, 15, 14, 13, 12, 11 from left to right; the tooth numbers in the second quadrant are 21, 22, 23, 24, 25, 26, 27, 28 from left to right; the tooth numbers in the third quadrant are 31, 32, 33, 34, 35, 36, 37, 38 from left to right; the tooth numbers in the fourth quadrant are 48, 47, 46, 45, 44, 43, 42, 41 from left to right.

[0045] Step S104: According to the second recognition model, recognize the abnormal teeth in the image to be processed, and obtain the second recognition result, where the second recognition result includes the second position information and image features of the abnormal teeth.

[0046] In this embodiment, after obtaining the image to be processed, it is also necessary to recognize the abnormal teeth in the image to be processed according to the second recognition model to obtain the second recognition result, that is, the second position information and image features of the abnormal teeth in the image to be processed. The image features are used to reflect features such as the color and shape of the abnormal teeth.

[0047] In an implementable manner, the second recognition model may be a convolutional neural network model. Specifically, the second recognition model may be a Faster R-CNN model. Faster R-CNN is a faster region-based convolutional neural network model, or it may be a YOLO (You Only Look Once) model. YOLO is a one-step object detection algorithm. The present disclosure does not limit the second recognition model.

[0048] Step S105: Generate a processing result of the dental panoramic tomography according to the tooth number information and the second recognition result.

[0049] In this embodiment, in order to generate the processing result of the dental panoramic tomography, it is necessary to bind the tooth number information and the second recognition result. Specifically, the tooth number information includes the tooth numbers and the first position information of all teeth, and the second recognition result includes the second position information and image features of the abnormal teeth. Binding the tooth number information and the second recognition result will obtain the processing result of the dental panoramic tomography, that is, [tooth number of the abnormal tooth, position information of the abnormal tooth, image features of the abnormal tooth]. If it is also necessary to record the recognition accuracy of the second recognition model, the processing result of the dental panoramic tomography may also be [tooth number of the abnormal tooth, position information of the abnormal tooth, image features of the abnormal tooth, recognition accuracy of the second recognition model].

[0050] In one implementable manner, the tooth labeling information can be bound to the second recognition result based on the relationship between the first position information and the second position information. Specifically, if the distance between the first position information and the second position information meets a preset threshold, it can be considered that the tooth corresponding to the first position information and the abnormal tooth corresponding to the second position information are the same tooth. At this time, the tooth labeling information of the tooth corresponding to the first position information can be bound to the second recognition result of the abnormal tooth corresponding to the second position information, so as to obtain the processing result of the dental panoramic image.

[0051] In the first embodiment of the present disclosure, first, the teeth in the image to be processed are recognized and labeled to obtain tooth labeling information, then the abnormal teeth in the image to be processed are recognized to obtain a second recognition result, and finally the tooth labeling information is bound to the second recognition result to generate the processing result of the dental panoramic image. In the first embodiment of the present disclosure, the information in the image to be processed is accurately extracted, and information such as the tooth label, position information, and image features of the abnormal teeth in the image to be processed are obtained, which can provide more useful information for doctors to judge tooth health and further reduce the burden on doctors.

[0052] Figure 3 It is a schematic flowchart of a method for processing a dental panoramic image according to the second embodiment of the present disclosure. As Figure 3 shown, before step S102, the method further includes:

[0053] Step S201, obtaining a first sample data set, where the first sample data set includes dental panoramic sample images.

[0054] In this embodiment, first, a number of dental panoramic sample images are obtained as the first sample data set, and then the dental panoramic sample images in the first sample data set can be preprocessed to improve the reliability of image recognition.

[0055] In one implementable manner, the dental panoramic sample images can be preprocessed such as grayscale conversion, geometric transformation, and image enhancement. Specifically, the component method can be used to perform grayscale conversion on the dental panoramic sample images; the OpenCV and bilinear interpolation algorithms are used to perform spatial transformation and scaling on the dental panoramic sample images respectively, where OpenCV is an open-source cross-platform computer vision and machine learning software library; the frequency domain method is used to enhance the dental panoramic sample images, and the present disclosure does not limit the preprocessing method.

[0056] Step S202, performing tooth annotation on the dental panoramic sample images in the first sample data set to obtain first training samples.

[0057] In this embodiment, after obtaining the first sample data set, tooth annotation can be directly performed on the panoramic dental sample images in the first sample data set, or tooth annotation can be performed on the preprocessed panoramic dental sample images, so as to obtain the first training sample.

[0058] In an implementable manner, the data annotation software LabelMe can be used to perform tooth annotation on the panoramic dental sample images in the first sample data set. First, a single tooth is framed and then annotated. A large number of teeth need to be annotated to ensure the accuracy of model recognition.

[0059] Step S203: Train the convolutional neural network according to the first training sample to obtain the first recognition model.

[0060] In this embodiment, the convolutional neural network can be trained according to the obtained first training sample to obtain the first recognition model.

[0061] In an implementable manner, the convolutional neural network can be the Mask Region Convolutional Neural Network (Mask R-CNN) based on a mask. Mask R-CNN can not only return the position information of the teeth but also return the binary mask of the teeth. The binary mask can be used to eliminate the information interference of non-tooth parts, further improving the usability of the data.

[0062] In the second embodiment of the present disclosure, the convolutional neural network is trained using the annotated panoramic dental sample images to obtain the first recognition model. The first recognition model can not only identify the positions of the teeth in the panoramic dental images but also return the binary masks of the teeth, enhancing the readability of the information.

[0063] Figure 4 It is a schematic flowchart of a method for processing panoramic dental images according to the third embodiment of the present disclosure. As Figure 4 shown, step S103 specifically includes:

[0064] Step S301: Group all the teeth in the image to be processed according to the first position information to obtain a grouping result.

[0065] In this embodiment, all the teeth in the image to be processed are numbered according to the first recognition result. In this embodiment, first, all the teeth in the image to be processed are grouped according to the first position information in the first recognition result to obtain a grouping result.

[0066] In an implementable embodiment, all teeth in the image to be processed can be divided into four groups according to the FDI tooth position notation, namely upper left (Quadrant I), upper right (Quadrant II), lower right (Quadrant III), and lower left (Quadrant IV). Specifically, the teeth in the image to be processed can be arranged and combined according to the abscissa and ordinate in the first position information to obtain the grouping result.

[0067] Step S302: Number all the teeth in the image to be processed according to the grouping result to obtain tooth numbering information.

[0068] In this embodiment, all the teeth in the image to be processed can be numbered according to the grouping result to obtain tooth numbering information. Specifically, the grouping result is used to characterize the positional relationship between each tooth and other teeth. Therefore, each tooth can be numbered according to the grouping result.

[0069] In an implementable embodiment, if the teeth in the image to be processed are grouped according to the FDI tooth position notation, the teeth in Quadrant I can be numbered from left to right as 18, 17, 16, 15, 14, 13, 12, 11; the teeth in Quadrant II can be numbered from left to right as 21, 22, 23, 24, 25, 26, 27, 28; the teeth in Quadrant III can be numbered from left to right as 31, 32, 33, 34, 35, 36, 37, 38; the teeth in Quadrant IV can be numbered from left to right as 48, 47, 46, 45, 44, 43, 42, 41. If the number of teeth in a certain quadrant is less than eight, the teeth can be numbered according to the number of teeth in the quadrant. That is, if there are six teeth in Quadrant I, the teeth in Quadrant I can be numbered from left to right as 16, 15, 14, 13, 12, 11. The tooth numbers and the first position information can be correspondingly stored in the data dictionary to obtain the tooth numbering information, that is, [tooth number: first position information].

[0070] In the third embodiment of the present disclosure, first, the teeth in the image to be processed are grouped, and then the teeth in the image to be processed are numbered according to the grouping result to obtain tooth numbering information. Numbering the teeth can uniquely determine the teeth, and the doctor can obtain the type and position information of the teeth according to the tooth numbers, which is convenient for the doctor to read the useful information in the panoramic dental image.

[0071] Figure 5 It is a schematic flowchart of a method for processing panoramic dental images according to the fourth embodiment of the present disclosure. As Figure 5 shown, step S301 specifically includes:

[0072] Step S401: Sort all the teeth in the image to be processed in ascending order according to the abscissa in the first position information to obtain the sorting result.

[0073] In this embodiment, all teeth in the image to be processed are grouped according to the first position information. In this embodiment, first, all teeth in the image to be processed need to be sorted in ascending order according to the abscissa in the first position information to obtain a sorting result. The abscissa in the first position information is used to represent the position of the tooth in the horizontal direction. Sorting all teeth in the image to be processed in ascending order according to the abscissa is equivalent to sorting the teeth in ascending order in the horizontal direction. The obtained sorting result can reflect the position of each tooth relative to other teeth in the horizontal direction.

[0074] Step S402: Calculate the ordinate centers of a specified number of teeth in sequence according to the sorting result.

[0075] In this embodiment, after obtaining the sorting result, the ordinate centers of a specified number of teeth can be calculated in sequence. The ordinate in the first position information is used to represent the position of the tooth in the vertical direction, and the ordinate center of the tooth is used to represent the ordinate of the center point of the height of the tooth detection frame.

[0076] In an implementable manner, the specified number is a preset rule. If the number of teeth in the image to be processed is greater than 6, the ordinate centers of 6 teeth are calculated in sequence; if the number of teeth in the image to be processed is less than 6, the ordinate centers of all teeth in the image to be processed are calculated. If there are 28 teeth in the image to be processed, then according to the sorting result, the ordinate centers of the 1st - 6th, 7th - 12th, 13th - 18th, and 19th - 24th teeth are calculated in sequence, and the remaining teeth are 4. At this time, the ordinate centers of the remaining 4 teeth are calculated; if there are only 4 teeth in the image to be processed, the ordinate centers of the 4 teeth in the image to be processed can be directly calculated.

[0077] In an implementable manner, if the first position information is [x, y, width, height], the calculation formula for the ordinate center is After obtaining the sorting result, the ordinate centers of 6 teeth can be calculated in sequence according to the above calculation formula for the ordinate center.

[0078] Figure 6 Shows a schematic diagram of the position information of a method for processing panoramic dental images according to the fourth embodiment of the present disclosure, as Figure 6 shown, the rectangle OABC is the image to be processed, and the rectangle DEFG is the tooth detection frame. Among them, the coordinates of point D are (x', y'), DE is the height height of the tooth detection frame, and DG is the width width of the tooth detection frame. Then the calculation formula for the ordinate center of the tooth corresponding to the rectangle DEFG is

[0079] Step S403: Calculate the mean value of the ordinate centers of a specified number of teeth.

[0080] In this embodiment, after calculating the ordinate centers of a specified number of teeth, it is necessary to calculate the mean value of the ordinate centers. Specifically, the mean value of the ordinate centers can be calculated by the method np.mean().

[0081] In an implementable manner, if the 1st to 6th teeth are selected from the sorting result to calculate the ordinate centers, the mean value of the ordinate centers of the 1st to 6th teeth is where P1 - P6 are the ordinate centers of the 1st to 6th teeth respectively.

[0082] Step S404: Group all the teeth in the image to be processed according to the ordinate and the mean value in the first position information, and obtain a grouping result.

[0083] In this embodiment, after obtaining the mean value of the ordinate centers of a specified number of teeth, it is necessary to group all the teeth in the image to be processed according to the ordinate and the mean value in the first position information, and obtain a grouping result.

[0084] In an implementable manner, all the teeth in the image to be processed can be first divided into an upper tooth group and a lower tooth group according to the ordinate and the mean value in the first position information, and then the teeth can be divided into a left upper tooth group, a right upper tooth group, a right lower tooth group, and a left lower tooth group according to the upper tooth group and the lower tooth group.

[0085] In the fourth embodiment of the present disclosure, all the teeth in the image to be processed are sorted in ascending order according to the abscissa in the first position information, then the ordinate centers of a specified number of teeth are calculated and the mean value is obtained according to the sorting result, and finally the teeth in the image to be processed are grouped according to the ordinate and the mean value in the first position information, so as to achieve precise grouping of the teeth in the image to be processed for subsequent labeling of the teeth.

[0086] Figure 7 It is a schematic flowchart of a method for processing a dental panoramic tomogram in the fifth embodiment of the present disclosure, as Figure 7 shown, step S404 specifically includes:

[0087] Step S501: Calculate the difference between the ordinate in the first position information and the mean value to obtain a first calculation result.

[0088] In this embodiment, after obtaining the mean value of the ordinate centers of a specified number of teeth, the difference between the ordinate in the first position information and the mean value can be calculated to obtain a first calculation result, and the first calculation result can indicate whether the tooth corresponding to the first position information is an upper tooth or a lower tooth.

[0089] In step S502, if the first calculation result is negative, the teeth corresponding to the ordinate in the first position information are classified into the upper tooth group.

[0090] In step S503, if the first calculation result is positive, the teeth corresponding to the ordinate in the first position information are classified into the lower tooth group.

[0091] In this embodiment, if the first calculation result is negative, that is, the ordinate in the first position information is less than the mean value, the teeth corresponding to the ordinate in the first position information are classified into the upper tooth group; if the first calculation result is positive, that is, the ordinate in the first position information is greater than the mean value, the teeth corresponding to the ordinate in the first position information are classified into the lower tooth group.

[0092] In an implementable manner, if the ordinate center points of the 1st - 6th teeth are calculated from the sorting result, and the mean value of the ordinate center points of the 1st - 6th teeth is m, if the ordinates in the first position information corresponding to the 1st, 3rd, and 5th teeth are less than the mean value m, then the 1st, 3rd, and 5th teeth are the upper tooth group; if the ordinates in the first position information corresponding to the 2nd, 4th, and 6th teeth are greater than the mean value m, then the 2nd, 4th, and 6th teeth are the lower tooth group.

[0093] In step S504, all the teeth in the image to be processed are grouped according to the upper tooth group and the lower tooth group to obtain a grouping result.

[0094] In this embodiment, the teeth can be classified into the upper left tooth group, the upper right tooth group, the lower right tooth group, and the lower left tooth group according to the upper tooth group and the lower tooth group, so as to obtain a grouping result.

[0095] In an implementable manner, the upper tooth group can be first divided into the upper left tooth group and the upper right tooth group, numbered according to the FDI tooth position representation method, then the midpoint between the teeth numbered 11 and 21 is taken as the division point, the teeth in the lower tooth group with abscissa less than the abscissa of the division point are classified into the lower left tooth group, the teeth in the lower tooth group with abscissa greater than the abscissa of the division point are classified into the lower right tooth group, and finally the lower left tooth group and the lower right tooth group are numbered according to the FDI tooth position representation method.

[0096] In the fifth embodiment of the present disclosure, the teeth in the image to be processed are classified into the upper tooth group and the lower tooth group by the difference between the ordinate in the first position information and the mean value, and then the upper tooth group and the lower tooth group are divided into the upper left tooth group, the upper right tooth group, the lower right tooth group, and the lower left tooth group, so as to obtain a grouping result, realizing the precise grouping of the teeth in the image to be processed for subsequent numbering of the teeth.

[0097] Figure 8 The flowchart of a method for processing a panoramic dental image according to the sixth embodiment of the present disclosure is shown, as Figure 8As shown, after numbering all the teeth in the image to be processed according to the first recognition result to obtain the tooth numbering information, the method further includes:

[0098] Step S601: Parse the binary masks of all the teeth in the image to be processed to obtain a binary tooth image.

[0099] In this embodiment, first, it is necessary to parse the binary masks of all the teeth in the image to be processed to obtain a binary tooth image. The binary tooth image can completely depict a single tooth in the image to be processed, eliminate the interference of information of non - tooth parts, and enhance the readability of the information.

[0100] Step S602: AND the binary tooth image with the image to be processed to obtain the original images of all the teeth in the image to be processed.

[0101] In this embodiment, the binary tooth image is a black - and - white image of the tooth contour. It is necessary to AND the binary tooth image with the image to be processed to obtain the original image of the tooth corresponding to the binary tooth image.

[0102] In an implementable manner, the image to be processed can be cropped according to the first position information to obtain the image of the tooth corresponding to the first position information. Then, the image of the tooth and the binary tooth image are cropped into pictures of the same size, and then the image of the tooth and the binary tooth image are ANDed to obtain the original image of the tooth corresponding to the binary tooth image.

[0103] Step S603: Perform background transparency processing on the original image to obtain an original image with a transparent background.

[0104] In this embodiment, after obtaining the original image of the tooth, it is necessary to perform background transparency processing on the original image to obtain an original image with a transparent background.

[0105] In an implementable manner, the RGB channels of the original image of the tooth can be converted to RGBA to obtain an alpha transparency channel. Among them, R, G, and B are the first letters of Red, Green, and Blue respectively, and A represents alpha. Alpha is the transparency channel, and the value of alpha can reflect the transparency of the corresponding pixel point. If the value of alpha is 255, the corresponding pixel point is black and completely opaque; if the value of alpha is 0, the corresponding pixel point is transparent. After converting the RGB channels of the original image of the tooth to RGBA, the pixel points with an alpha value of 255 can be traversed in a loop and set to 0, so as to obtain an original image with a transparent background.

[0106] Step S604: Name and save the original image with a transparent background with its corresponding tooth number.

[0107] In this embodiment, after obtaining the original image with a transparent background, it can be named and saved with the corresponding tooth label. By naming the original image with a transparent background with the corresponding tooth label, the original image with a transparent background of the corresponding tooth can be retrieved according to the tooth label, facilitating the display of the original image with a transparent background of the tooth.

[0108] In the sixth embodiment of the present disclosure, the binary tooth image is ANDed with the image to be processed to obtain the original image of the tooth, and the original image of the tooth is subjected to background transparency processing. Then, the original image with a transparent background is named and saved with the corresponding tooth label, which not only improves the readability of the original image of the tooth but also facilitates the display of the original image with a transparent background of the tooth.

[0109] Figure 9 It is a schematic flowchart of a method for processing panoramic dental images according to the seventh embodiment of the present disclosure. As Figure 9 shown, step S104 specifically includes:

[0110] Step S701: Extract features from the image to be processed according to the second recognition model to obtain the feature map of the image to be processed.

[0111] In this embodiment, the second recognition model is used to identify abnormal teeth in the image to be processed. First, features are extracted from the image to be processed according to the second recognition model to obtain the feature map of the image to be processed. Specifically, if the second recognition model is a convolutional neural network, the second recognition model can perform convolution on the image to be processed to extract the feature map of the image to be processed.

[0112] Step S702: Locate the abnormal teeth in the image to be processed according to the feature map of the image to be processed to obtain the second position information of the abnormal teeth.

[0113] In this embodiment, the abnormal teeth in the image to be processed can be identified according to the feature map of the image to be processed, and the second position information of the abnormal teeth is returned. The abnormal teeth are teeth with significant differences in shape, color, etc. from normal teeth.

[0114] Step S703: Output the second position information of the abnormal teeth and the image features of the abnormal teeth corresponding to the second recognition model.

[0115] In this embodiment, after identifying the abnormal teeth in the image to be processed, the second position information of the abnormal teeth and the image features of the abnormal teeth corresponding to the second recognition model can be output.

[0116] In an implementable embodiment, the second recognition model is used to recognize abnormal teeth of a specific shape or color, such as a second recognition model for recognizing teeth with severely worn shapes or a second recognition model for recognizing teeth with a darker color. After the second recognition model recognizes an abnormal tooth, it is necessary to output not only the second position information of the abnormal tooth but also the image features of the abnormal tooth corresponding to the second recognition model that recognized the abnormal tooth.

[0117] In the seventh embodiment of the present disclosure, the second recognition model is used to recognize abnormal teeth in the image to be processed, obtaining the position information and image features of the abnormal teeth, so as to bind the tooth numbering information to the second recognition result later and facilitate the doctor to judge the health status of the teeth according to the image features of the abnormal teeth.

[0118] Figure 10 The flowchart of a method for processing panoramic dental images according to the eighth embodiment of the present disclosure is shown, as Figure 10 shown, step S105 specifically includes:

[0119] Step S801, calculate the distances between the second position information of the abnormal teeth and the first position information of all teeth in the image to be processed in sequence, obtaining a second calculation result.

[0120] In this embodiment, the tooth numbering information includes the first position information of the teeth in the image to be processed, and the second recognition result includes the second position information of the abnormal teeth in the image to be processed. First, it is necessary to calculate the distances between the second position information of the abnormal teeth and the first position information of all teeth in the image to be processed, obtaining a second calculation result.

[0121] In an implementable embodiment, the distance between the central coordinates of the first position information and the second position information can be calculated to obtain a second calculation result. If the first position information is [x, y, width, height] and the second position information is [x2, y2, width2, height2], then the central coordinate of the first position information is The central coordinate of the second position information is The distance between the central coordinate of the first position information and the central coordinate of the second position information can be calculated according to the Euclidean distance.

[0122] Step S802, select the minimum distance between the second position information of the abnormal teeth and the first position information of all teeth in the image to be processed according to the second calculation result.

[0123] In this embodiment, it is necessary to select the minimum distance between the second position information of the abnormal tooth in the second calculation result and the first position information of all teeth in the image to be processed. If the recognition accuracies of the first recognition model and the second recognition model reach the preset threshold, the second position information of the abnormal tooth and the first position information corresponding to the minimum distance should be the position information of the same tooth.

[0124] Step S803: Bind the second recognition result to the tooth label information corresponding to the minimum distance to generate a processing result of the dental panoramic image.

[0125] In this embodiment, after obtaining the minimum distance, the second recognition result can be bound to the tooth label information corresponding to the minimum distance to generate a processing result of the dental panoramic image. Since the second position information of the abnormal tooth and the first position information corresponding to the minimum distance should be the position information of the same tooth, the tooth label information corresponding to the first position information and the second recognition result are the information of the same tooth. Therefore, they can be bound to obtain a complete processing result of the dental panoramic image.

[0126] In an implementable manner, if the tooth label information corresponding to the first position information is [tooth label, first position information, original image with a transparent background], and the second recognition result is [second position information of the abnormal tooth, recognition accuracy of the second recognition model, image features of the abnormal tooth], then the processing result of the dental panoramic image obtained by binding is [tooth label, position information, original image with a transparent background, recognition accuracy of the second recognition model, image features of the abnormal tooth].

[0127] In the eighth embodiment of the present disclosure, the second recognition result is bound to the tooth label information corresponding to the minimum distance to generate a processing result of the dental panoramic image, maximizing the extraction of useful information from the dental panoramic image for the doctor to judge the dental health condition, further improving the judgment efficiency and the accuracy of the judgment result.

[0128] Figure 11 It is a schematic structural diagram of a dental panoramic image processing device according to the ninth embodiment of the present disclosure, as Figure 9As shown in the figure, the device mainly includes: an acquisition module 90, configured to acquire an image to be processed, where the image to be processed is a panoramic dental image; a first recognition module 91, configured to recognize the teeth in the image to be processed according to a first recognition model, and obtain a first recognition result, where the first recognition result includes the first position information of all teeth in the image to be processed and a binary mask; a labeling module 92, configured to label all teeth in the image to be processed according to the first recognition result, and obtain tooth labeling information, where the tooth labeling information includes tooth labels and the first position information of the teeth corresponding to the tooth labels; a second recognition module 93, configured to recognize abnormal teeth in the image to be processed according to a second recognition model, and obtain a second recognition result, where the second recognition result includes the second position information and image features of the abnormal teeth; and a generation module 94, configured to generate a panoramic dental image processing result according to the tooth labeling information and the second recognition result.

[0129] In an implementable embodiment, the device further includes: a second acquisition module, configured to acquire a first sample data set, where the first sample data set includes panoramic dental sample images; an annotation module, configured to perform tooth annotation on the panoramic dental sample images in the first sample data set, and obtain first training samples; and a training module, configured to train a convolutional neural network according to the first training samples, and obtain a first recognition model.

[0130] In an implementable embodiment, the labeling module 92 includes: a grouping sub-module, configured to group all teeth in the image to be processed according to the first position information, and obtain a grouping result; and a labeling sub-module, configured to label all teeth in the image to be processed according to the grouping result, and obtain tooth labeling information.

[0131] In an implementable embodiment, the grouping sub-module specifically includes: a sorting unit, configured to sort all teeth in the image to be processed in ascending order according to the abscissa in the first position information, and obtain a sorting result; a first calculation unit, configured to sequentially calculate the ordinate center points of a specified number of teeth according to the sorting result; a second calculation unit, configured to calculate the mean value of the ordinate center points of the specified number of teeth; and a grouping unit, configured to group all teeth in the image to be processed according to the ordinate and the mean value in the first position information, and obtain a grouping result.

[0132] In one possible implementation, the grouping unit specifically includes: a calculation subunit, configured to calculate the difference between the ordinate in the first position information and the mean value to obtain a first calculation result; a first grouping subunit, configured to, if the first calculation result is negative, classify the teeth corresponding to the ordinate in the first position information into the upper tooth group; a second grouping subunit, configured to, if the first calculation result is positive, classify the teeth corresponding to the ordinate in the first position information into the lower tooth group; and a third grouping subunit, configured to group all the teeth in the image to be processed according to the upper tooth group and the lower tooth group to obtain a grouping result.

[0133] In one possible implementation, the apparatus further includes: an analysis module, configured to analyze the binary mask of all the teeth in the image to be processed to obtain a tooth binary map; an AND module, configured to perform an AND operation on the tooth binary map and the image to be processed to obtain the original image of all the teeth in the image to be processed; a transparency processing module, configured to perform background transparency processing on the original image to obtain an original image with a transparent background; and a saving module, configured to name and save the original image with a transparent background according to the corresponding tooth label.

[0134] In one possible implementation, the second recognition module 93 specifically includes: a feature extraction submodule, configured to extract features from the image to be processed according to the second recognition model to obtain a feature map of the image to be processed; a positioning submodule, configured to position the abnormal teeth in the image to be processed according to the feature map of the image to be processed to obtain the second position information of the abnormal teeth; and an output submodule, configured to output the second position information of the abnormal teeth and the image features of the abnormal teeth corresponding to the second recognition model.

[0135] In one possible implementation, the generation module 94 specifically includes: a calculation submodule, configured to sequentially calculate the distances between the second position information of the abnormal teeth and the first position information of all the teeth in the image to be processed to obtain a second calculation result; a selection submodule, configured to select the minimum distance between the second position information of the abnormal teeth and the first position information of all the teeth in the image to be processed according to the second calculation result; and a generation submodule, configured to bind the second recognition result to the tooth label information corresponding to the minimum distance to generate a processing result of the dental panoramic tomography image.

[0136] According to an embodiment of the present disclosure, the present disclosure further provides an electronic device and a readable storage medium.

[0137] Figure 12FIG. shows a schematic block diagram of an exemplary electronic device 1200 that can be used to implement embodiments of the present disclosure. The electronic device is intended to represent various forms of digital computers, such as, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, personal digital processors, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely exemplary and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0138] As Figure 12 shown, the device 1200 includes a computing unit 1201 that can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 1202 or a computer program loaded from a storage unit 1208 into a random access memory (RAM) 1203. In the RAM 1203, various programs and data required for the operation of the device 1200 can also be stored. The computing unit 1201, the ROM 1202, and the RAM 1203 are connected to each other via a bus 1204. An input / output (I / O) interface 1205 is also connected to the bus 1204.

[0139] A plurality of components in the device 1200 are connected to the I / O interface 1205, including: an input unit 1206, such as a keyboard, a mouse, etc.; an output unit 1207, such as various types of displays, speakers, etc.; a storage unit 1208, such as a magnetic disk, an optical disk, etc.; and a communication unit 1209, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 1209 allows the device 1200 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0140] The computing unit 1201 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 1201 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 1201 executes the various methods and processes described above, such as a method for processing panoramic dental images. For example, in some embodiments, a method for processing panoramic dental images can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as the storage unit 1208. In some embodiments, part or all of the computer program can be loaded and / or installed onto the device 1200 via the ROM 1202 and / or the communication unit 1209. When the computer program is loaded into the RAM 1203 and executed by the computing unit 1201, one or more steps of a method for processing panoramic dental images described above can be executed. Alternatively, in other embodiments, the computing unit 1201 can be configured to execute a method for processing panoramic dental images by any other suitable means (e.g., by means of firmware).

[0141] The various embodiments of the systems and techniques described above in this document can be implemented in digital electronic circuitry, integrated circuit systems, field-programmable gate arrays (FPGA), application-specific integrated circuits (ASIC), application-specific standard products (ASSP), systems-on-chip (SOC), complex programmable logic devices (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special-purpose or general-purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0142] The program code for implementing the methods of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the program codes are executed by the processor or controller, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The program codes can be executed entirely on the machine, partially on the machine, as an independent software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0143] In the context of this disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of the machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, 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 disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0144] To provide for interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can also be used to provide for interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic, speech, or tactile input).

[0145] The systems and techniques described herein can be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0146] A computer system may include a client and a server. The client and the server are generally far from each other and usually interact via a communication network. The relationship between the client and the server is generated by computer programs running on corresponding computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server incorporating a blockchain.

[0147] It should be understood that various forms of processes shown above can be used, steps can be reordered, added or deleted. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0148] In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one such feature. In the description of this disclosure, "a plurality of" means two or more, unless otherwise specifically defined.

[0149] As described above, this is only the specific implementation manner of this disclosure, but the protection scope of this disclosure is not limited thereto. Any person skilled in the art within the technical scope disclosed in this disclosure can easily think of changes or substitutions, which should all be covered within the protection scope of this disclosure. Therefore, the protection scope of this disclosure should be subject to the protection scope of the claims.

Claims

1. A method for processing panoramic dental images, characterized in that, The method includes: Obtaining a to-be-processed image, where the to-be-processed image is a panoramic dental image; Identifying teeth in the to-be-processed image according to a first recognition model to obtain a first recognition result, where the first recognition result includes first position information of all teeth in the to-be-processed image and a binary mask; Numbering all teeth in the to-be-processed image according to the first recognition result to obtain tooth numbering information, where the tooth numbering information includes tooth numbers and first position information of teeth corresponding to the tooth numbers; Identifying abnormal teeth in the to-be-processed image according to a second recognition model to obtain a second recognition result, where the second recognition result includes second position information of the abnormal teeth and image features; Generating a panoramic dental image processing result according to the tooth numbering information and the second recognition result; Among them, the identifying abnormal teeth in the to-be-processed image according to the second recognition model to obtain a second recognition result includes: Extracting features of the to-be-processed image according to the second recognition model to obtain a feature map of the to-be-processed image; Locating abnormal teeth in the to-be-processed image according to the feature map of the to-be-processed image to obtain second position information of the abnormal teeth; Outputting the second position information of the abnormal teeth and image features of the abnormal teeth corresponding to the second recognition model; Among them, the generating a panoramic dental image processing result according to the tooth numbering information and the second recognition result includes: Calculating distances between the second position information of the abnormal teeth and the first position information of all teeth in the to-be-processed image in sequence to obtain a second calculation result; Selecting the minimum distance between the second position information of the abnormal teeth and the first position information of all teeth in the to-be-processed image according to the second calculation result; Binding the second recognition result to the tooth numbering information corresponding to the minimum distance to generate the panoramic dental image processing result, where the panoramic dental image processing result includes tooth numbers, position information, an original image with a transparent background, the recognition accuracy of the second recognition model, and image features of abnormal teeth.

2. The method according to claim 1, wherein Before the identifying teeth in the to-be-processed image according to the first recognition model to obtain a first recognition result, the method further includes: Obtaining a first sample dataset, where the first sample dataset includes panoramic dental sample images; Performing tooth annotation on the panoramic dental sample images in the first sample dataset to obtain first training samples; Training a convolutional neural network according to the first training samples to obtain the first recognition model.

3. The method according to claim 1, characterized in that, The numbering all teeth in the to-be-processed image according to the first recognition result to obtain tooth numbering information includes: Grouping all teeth in the to-be-processed image according to the first position information to obtain a grouping result; Numbering all teeth in the to-be-processed image according to the grouping result to obtain the tooth numbering information.

4. The method according to claim 3, characterized in that, The grouping all teeth in the to-be-processed image according to the first position information to obtain a grouping result includes: Sort all the teeth in the image to be processed in ascending order according to the abscissa in the first position information to obtain a sorting result; Calculate the vertical center points of a specified number of teeth in sequence according to the sorting result; Calculate the average value of the vertical center points of the specified number of teeth; Group all the teeth in the image to be processed according to the ordinate in the first position information and the average value to obtain a grouping result.

5. The method according to claim 4, wherein The step of grouping all the teeth in the image to be processed according to the ordinate in the first position information and the average value to obtain a grouping result includes: Calculate the difference between the ordinate in the first position information and the average value to obtain a first calculation result; If the first calculation result is negative, classify the teeth corresponding to the ordinate in the first position information into the upper tooth group; If the first calculation result is positive, classify the teeth corresponding to the ordinate in the first position information into the lower tooth group; Group all the teeth in the image to be processed according to the upper tooth group and the lower tooth group to obtain a grouping result.

6. The method according to claim 1, wherein After numbering all the teeth in the image to be processed according to the first recognition result to obtain tooth numbering information, the method further includes: Parse the binary masks of all the teeth in the image to be processed to obtain a tooth binary image; Perform an AND operation on the tooth binary image and the image to be processed to obtain the original images of all the teeth in the image to be processed; Perform background transparency processing on the original images to obtain original images with a transparent background; Save the original images with a transparent background with the corresponding tooth numbers as their names.

7. An intraoral panoramic image processing apparatus, characterized in that, The device includes: An acquisition module for acquiring an image to be processed, where the image to be processed is a dental panoramic tomography image; A first recognition module for recognizing the teeth in the image to be processed according to a first recognition model to obtain a first recognition result, where the first recognition result includes the first position information and binary masks of all the teeth in the image to be processed; A numbering module for numbering all the teeth in the image to be processed according to the first recognition result to obtain tooth numbering information, where the tooth numbering information includes tooth numbers and the first position information of the teeth corresponding to the tooth numbers; A second recognition module for recognizing abnormal teeth in the image to be processed according to a second recognition model to obtain a second recognition result, where the second recognition result includes the second position information and image features of the abnormal teeth; A generation module for generating a dental panoramic tomography image processing result according to the tooth numbering information and the second recognition result; Among them, the step of recognizing abnormal teeth in the image to be processed according to the second recognition model to obtain a second recognition result includes: Extract features of the image to be processed according to the second recognition model to obtain a feature map of the image to be processed; Locate the abnormal teeth in the image to be processed according to the feature map of the image to be processed to obtain the second position information of the abnormal teeth; Output the second position information of the abnormal teeth and the image features of the abnormal teeth corresponding to the second recognition model; Among them, generating the processing result of the dental panoramic image according to the tooth label information and the second recognition result includes: Calculating the distances between the second position information of the abnormal tooth and the first position information of all teeth in the image to be processed in sequence to obtain a second calculation result; Selecting the minimum distance between the second position information of the abnormal tooth and the first position information of all teeth in the image to be processed according to the second calculation result; Binding the second recognition result with the tooth label information corresponding to the minimum distance to generate the processing result of the dental panoramic image, where the processing result of the dental panoramic image includes tooth label, position information, the original image with a transparent background, the recognition accuracy of the second recognition model, and the image features of the abnormal tooth.

8. An electronic device, characterized in that, It includes: At least one processor; And A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method according to any one of claims 1-6.

9. A non-transitory computer-readable storage medium storing computer instructions, characterized in that, The computer instructions are used to cause the computer to execute the method according to any one of claims 1-6.

Images