An intelligent image detection method for orthodontic treatment

Abstract

The present invention discloses an intelligent image detection method for oral orthodontic treatment, which relates to the technical field of oral orthodontic image detection. It is implemented by a data acquisition module, a data processing module, an image processing module and a result output application module, and the pre-processed image data is converted into a gray space, and the image processing module is used to process the image data. The present invention has the introduction and weighted calculation based on the three channels of red, green and blue, as well as the analysis and comprehensive calculation of gray values, original color images and contrast, so as to retain part of the color information and reduce information loss, improve the contrast and clarity of the gray image, make the boundaries of teeth and periodontal tissues clearer, and realize more accurate and comprehensive and reliable orthodontic effect evaluation. In addition, the adjacent submodules of the present method are associated and the overall algorithm system has the characteristics of cyclic drop, so that the practicality of the present method is better and the adaptability is stronger.

Description

Technical Field

[0001] The present invention relates to the technical field of orthodontic image detection, and in particular to an intelligent image detection method for orthodontic treatment. Background Art

[0002] Orthodontic treatment is a measure to correct teeth and eliminate malocclusion and deformity. Orthodontic treatment can achieve the effect of beautiful teeth. During orthodontic treatment, medical staff need to obtain the patient's oral condition. With the popularization of 3D scanning and graphic image processing, the field of orthodontics has also ushered in a digital revolution. The existing intelligent image detection for orthodontic treatment obtains the 3D point cloud model of the patient's jaw through 3D scanning equipment and then performs image processing to facilitate medical staff to observe the inside of the mouth.

[0003] However, in the current existing orthodontic treatment image detection methods, it may be difficult to fully utilize all the image information during image processing, especially when there are large external interference factors, such as dental images with complex lighting conditions and color changes, and when directly converting color images to grayscale images, key color information may be lost, thereby affecting the accuracy of subsequent image analysis and detection. Therefore, the existing image detection methods may not be effective in extracting key features such as the boundaries and morphology of teeth and periodontal tissues. Summary of the invention

[0004] The purpose of the present invention is to provide an intelligent image detection method for oral orthodontic treatment, which solves the problems raised in the above-mentioned background technology.

[0005] To achieve the above object, the present invention provides the following technical solution: an intelligent image detection method for oral orthodontic treatment, comprising the following image detection steps:

[0006] The first step is to scan the patient's oral cavity through the data acquisition module to obtain a three-dimensional point cloud model of the patient's jaws, and extract two-dimensional slice image data from the three-dimensional point cloud data;

[0007] In the second step, the two-dimensional slice image data is preprocessed by the data processing module, and the initial weights of the red, green and blue channels are set;

[0008] The third step is to convert the preprocessed two-dimensional slice image data into grayscale space, and obtain the initial grayscale value, image fusion value and image feature comprehensive value through the image processing module;

[0009] In the fourth step, the initial grayscale value, image fusion value and image feature comprehensive value are displayed on the user interface through the result output application module for the formulation and evaluation of oral orthodontic treatment plans.

[0010] Optionally, the image processing module includes a grayscale submodule, a gray-color synthesis submodule, an image detection synthesis result submodule and an adjustment submodule;

[0011] The processing process of the image processing module includes:

[0012] S1: Input the red channel value, green channel value and blue channel value into the grayscale submodule and output the initial grayscale value;

[0013] S2: Input the initial grayscale value and contrast value into the gray-color synthesis submodule and output the image fusion value;

[0014] S3: input the initial gray value and the image fusion value into the image feature comprehensive value, and output the image feature comprehensive value;

[0015] S4: Input the comprehensive value of image features into the adjustment submodule, output a new red channel variable weight, replace the original red channel variable weight in the grayscale submodule with the new red channel variable weight to optimize the initial grayscale value, continuously adjust the initial grayscale value through iteration, set the number of iterations to six, and initialize the iteration counter.

[0016] Optionally, the processing process of the grayscale submodule is as follows:

[0017] OU=(PA+PB+PC) 2 *B 1 +B 2 ;

[0018] PA=A 1 *a 1 ;

[0019] PB=A 2 *b;

[0020] PC=A 3 *c;

[0021] in:

[0022] OU is the initial gray value;

[0023] A 1 is the red channel value;

[0024] A 2 is the green channel value;

[0025] A 3 is the blue channel value;

[0026] a 1 Variable weight for the red channel;

[0027] b is the green channel weight;

[0028] c is the blue channel weight;

[0029] B 1 is the grayscale scaling factor value;

[0030] B 2 is the grayscale value offset value;

[0031] PA is the red channel contribution value;

[0032] PB is the green channel contribution value;

[0033] PC is the contribution value of the blue channel.

[0034] Through the red channel variable weight a 1 , green channel weight b and blue channel weight c control the red channel value A 1 , Green channel value A 2 , blue channel value A 3 The contribution values ​​of the three color channels to highlight the features of the teeth and jaws in the grayscale image, as well as the grayscale value scaling factor value B 1 And gray value offset value B 2 The grayscale value range and offset are adjusted to optimize the brightness and contrast of the grayscale image, and the initial grayscale value OU is output.

[0035] Optionally, the processing process of the gray color comprehensive submodule is as follows:

[0036] OW=d*OU+e*(A 1 +A 2 +A 3 )+PD;

[0037] PD = f * C;

[0038] in:

[0039] OW is the image fusion value;

[0040] d is the gray value weight factor;

[0041] e is A 1 , A 2 and A 3 The weight factor of the sum;

[0042] f is the contrast weight factor;

[0043] C is the contrast value;

[0044] PD is the contrast contribution value;

[0045] The contrast contribution value PD is combined with the initial grayscale value OU output by the grayscale submodule to enhance the contrast of the fused image and output the image fusion value OW.

[0046] Optionally, the processing process of the image detection comprehensive result submodule is as follows:

[0047] OP = g*OU+OW*h+D;

[0048] UA=g*OU

[0049] in:

[0050] OP is the comprehensive value of image features;

[0051] g is the initial gray value weight factor;

[0052] h is the weight factor of the fusion result;

[0053] D is the texture feature value of the image;

[0054] UA is the comprehensive contribution value of the initial gray value;

[0055] Based on the initial grayscale value OU output by the grayscale submodule and the image fusion value OW output by the gray-color synthesis submodule, the contribution of the initial grayscale value OU and the image fusion value OW in OP is controlled by adjusting the initial grayscale value weight factor g and the fusion result weight factor h to balance the initial grayscale value OU and the image fusion value OW, and output the image feature comprehensive value OP.

[0056] Optionally, the process of adjusting the submodule is as follows:

[0057] According to the new red channel variable weight a n =a 1 +0.01*OP;

[0058] If a n >1, then output a n is 1, and a n Replace a in the grayscale submodule 1 ;

[0059] If a n <0, then output a n is 0, and a n Replace a in the grayscale submodule 1 ;

[0060] If 0 n ≤1, then a n Normal output, a n Replace a in the grayscale submodule 1 ; ​

[0061] The image feature comprehensive value OP is input to the adjustment submodule, and the new red channel variable weight a is output. n , set the new red channel variable weight a n Replace the original red channel variable weight a in the grayscale submodule 1 , to optimize the initial gray value OU, so as to continuously adjust the initial gray value OU through iteration, and set the number of iterations to six times, and make the original red channel variable weight a 1 Make adjustments to highlight teeth and jaw features and improve the accuracy of orthodontic image detection.

[0062] Optionally, the devices used by the output and application module are a display, a printer and treatment planning software.

[0063] Optionally, the devices used by the data acquisition module are three-dimensional scanning equipment and image processing software.

[0064] Compared with the prior art, the present invention has the following beneficial effects:

[0065] 1. The present invention obtains the initial grayscale value through the grayscale submodule. The calculation of the initial grayscale value is convenient for users to analyze and use, so as to improve the contrast and clarity of the grayscale image, make the boundary between teeth and periodontal tissues clearer, and facilitate subsequent feature extraction and detection.

[0066] 2. The present invention obtains an image fusion value through a gray-color integrated sub-module. The image fusion value includes an image of grayscale information and original color information, so that the fused image has both clear boundaries of the grayscale image and color information of the color image, providing a richer information source for subsequent image analysis and detection, and improving the accuracy of detection.

[0067] 3. The present invention obtains the image feature comprehensive value through the image detection comprehensive result sub-module. The application of the image feature comprehensive value integrates a variety of image information and features, making the detection result more accurate and comprehensive, and reducing the decision analysis error caused by inaccurate feature extraction.

[0068] 4. The present invention affects the variable weight of the red channel in the grayscale submodule by adjusting the form of submodule iteration, thereby affecting the initial grayscale value. Since the red channel has higher sensitivity or contrast in actual application, the structural information of the teeth and jaws is more prominent in the image through continuous automatic iterative adjustment of the variable weight of the red channel, and the contrast with the surrounding tissues is enhanced, making the details of these key areas clearer and more discernible. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 A flowchart of the steps of the intelligent image detection method for orthodontic treatment is provided;

[0070] Figure 2 The overall structure diagram of the intelligent image detection method for orthodontic treatment is shown in FIG.

[0071] Figure 3 This is a schematic diagram of the overall structure of the image processing module of the intelligent image detection method for orthodontic treatment;

[0072] Figure 4 This is a schematic diagram of the overall structure of the data acquisition module of the intelligent image detection method for orthodontic treatment. DETAILED DESCRIPTION

[0073] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0074] Regarding the present intelligent image detection method for oral orthodontic treatment, it is different from the existing orthodontic treatment image detection methods. The existing methods often cannot make full use of the red channel values, green channel values ​​and blue channel values, especially when processing dental images with complex lighting conditions and color changes, and directly converting color images into grayscale images may lose key color information, which is crucial for accurately identifying the boundaries of teeth and periodontal tissues, and may affect the accuracy of subsequent image analysis and detection. The present intelligent image detection method for oral orthodontic treatment is based on the introduction of grayscale, channel values, contrast and other additional numerical values, so that the three algorithm units form an overall iterative system, thereby achieving more accurate, comprehensive and reliable orthodontic effect evaluation, which is of great significance for improving the quality and efficiency of orthodontic treatment, and helps doctors to formulate more scientific and reasonable orthodontic treatment plans.

[0075] For examples, see Figures 1 to 4 ,This implementation provides an intelligent image detection method for oral orthodontic treatment, which is implemented by a data acquisition module, a data processing module, an image processing module and a result output application module;

[0076] The specific implementation process is as follows:

[0077] The patient's oral cavity is scanned by the data acquisition module, specifically using an intraoral scanner to obtain a 3D point cloud model of the patient's jaws. Appropriate software is selected to process the 3D point cloud data to obtain sliced ​​2D point cloud data. Commonly used software includes PCL (Point Cloud Library), MATLAB, CloudCompare, etc. The sliced ​​2D point cloud data is converted into an image format. The generated 2D slice image is further processed using image processing software (such as Photoshop, GIMP, etc.) or programming libraries (such as OpenCV) to obtain 2D slice image data, including the red channel value A. 1 , Green channel value A 2 and the blue channel value A 3 ;

[0078] Preprocess the two-dimensional slice image data through the data processing module and set the initial weights of the red, green and blue channels;

[0079] The preprocessed two-dimensional slice image data is converted into grayscale space, and the image processing module is used to obtain the initial grayscale value, image fusion value and image feature comprehensive value;

[0080] The initial grayscale value, image fusion value and image feature comprehensive value are displayed on the user interface through the result output application module for the formulation and evaluation of orthodontic treatment plans;

[0081] Furthermore, the image processing module includes a grayscale submodule, a gray-color synthesis submodule, an image detection synthesis result submodule and an adjustment submodule.

[0082] In this embodiment, the method forms an intelligent image detection system for oral orthodontic treatment through three interrelated and intertwined sub-modules, which can achieve more accurate, comprehensive and reliable orthodontic effects. The evaluation and detection results are of great significance to improving the quality and efficiency of orthodontic treatment, and help doctors to formulate more scientific and reasonable orthodontic treatment plans. The three sub-modules are combined to fully consider the role of grayscale information and color information in oral orthodontic image detection, and through flexible adjustment of weights, it is ensured that the contribution of different information in the final result is optimized to improve the comprehensiveness and accuracy of the detection. From the initial grayscale value calculation to the image fusion value calculation and analysis, and then to the image feature comprehensive value, the three sub-modules jointly optimize the brightness, contrast and detail performance of the image, making the image more suitable for the analysis and evaluation of orthodontic treatment, and the overall algorithm system has high flexibility and scalability, and can adapt to changes in demand in different application scenarios;

[0083] By integrating grayscale information and color information, the image detection system can more accurately evaluate the arrangement of teeth, such as the inclination angle of the teeth, the degree of crowding, etc., and the overall image detection system can comprehensively consider the shape, position and relative relationship of the jaw with other structures, and provide doctors with accurate information about the position of the jaw. Therefore, the image detection system can more accurately predict the effect of orthodontic treatment and provide a reference for doctors to formulate treatment plans.

[0084] See also Figures 1 to 4 , the processing process of the grayscale submodule is as follows:

[0085] OU=(PA+PB+PC) 2 *B 1 +B 2 ;

[0086] PA=A 1 *a 1 , PB=A 2 *b, PC = A 3 *c;

[0087] in:

[0088] OU is the initial gray value;

[0089] A 1 is the red channel value, A 2 is the green channel value, A 3 is the blue channel value;

[0090] a 1 Variable weight for the red channel;

[0091] b is the green channel weight;

[0092] c is the blue channel weight;

[0093] B 1 is the grayscale scaling factor value;

[0094] B 2 is the grayscale value offset value;

[0095] PA is the red channel contribution value;

[0096] PB is the green channel contribution value;

[0097] PC is the contribution value of the blue channel.

[0098] In this embodiment, for the multi-color channel input adopted by this submodule, since the color of the tissue in the cavity is relatively single, mainly pink or red, in digital image processing, even images that appear to be single-tone will be captured and processed in the form of red, green and blue color space. This is because the image capture device usually uses red, green and blue to record the color information. The calculation is based on the three-color channel value combined with B 1 and B 2 This initial grayscale value reflects the grayscale representation of the image in a specific weighted manner, and is obtained by a 1 Under the influence of the weights of a, b, and c, the contribution of the red, green, and blue color channels in grayscale calculation can be flexibly controlled, which helps to adapt to the different requirements of grayscale images in different application scenarios. For example, if a color channel contains important information in the image, the performance of this information in the grayscale image can be enhanced by increasing the weight of the channel. 1 and B 2 The grayscale value range and offset can be adjusted to ensure that the grayscale image has appropriate brightness and contrast within a specific range;

[0099] Different from the existing grayscale calculation methods, such as the average method and the maximum method, which do not consider the weight difference between color channels and therefore cannot flexibly adjust the color contribution, this method achieves more flexible and accurate grayscale calculation by introducing weight factors through the weighted grayscale calculation method. This submodule can highlight the features of teeth and jaws in grayscale images by adjusting the weights of the red, green, and blue color channels. 1 and B 2 Adjustment of parameters can optimize the brightness and contrast of grayscale images, making the details of teeth and jaws clearer and facilitating subsequent orthodontic effect evaluation.

[0100] See also Figures 1 to 4 ,The processing process of the gray color comprehensive submodule is as follows:

[0101] OW=d*OU+e*(A 1 +A 2 +A 3 )+PD;

[0102] PD = f * C;

[0103] in:

[0104] OW is the image fusion value;

[0105] d is the gray value weight factor;

[0106] e is A 1 , A 2 and A 3 The weight factor of the sum;

[0107] f is the contrast weight factor;

[0108] C is the contrast value;

[0109] PD is the contrast contribution value.

[0110] In this embodiment, this submodule performs comprehensive calculation based on the calculation result of the grayscale submodule and the values ​​involved in the grayscale submodule calculation, and uses contrast to generate a comprehensive processing result value, which can reflect the characteristics and properties of the image after the grayscale and color information are fused;

[0111] This submodule can be expanded from a simple grayscale image to an image that incorporates color information, so that the image contains more dimensional information, which is beneficial to subsequent orthodontic image detection. By fusing the grayscale image with the color image, the details and color information of the color image can be introduced while retaining the overall brightness and contrast information of the grayscale image. The fused image can be contrast enhanced through PD to further improve the visual effect of the image. Different from the existing simple grayscale image and color image superposition form, this submodule introduces the initial grayscale value, the three-color channel value, and the contrast-related numerical value to achieve more refined image fusion and effect enhancement. Compared with the multi-scale fusion method, this algorithm unit may have more advantages in computational efficiency and real-time performance. It does not require complex operations such as multi-scale decomposition and reconstruction. Combining the overall brightness and contrast information of the grayscale image with the color and detail information of the color image can generate richer and more accurate orthodontic images, which is crucial for evaluating the orthodontic treatment effects such as tooth arrangement and jaw position.

[0112] See also Figures 1 to 4 ,The processing process of the image detection comprehensive result submodule is as follows:

[0113] OP=g*OU+OW*h+D

[0114] UA=g*OU

[0115] in:

[0116] OP is the comprehensive value of image features;

[0117] g is the initial gray value weight factor;

[0118] h is the weight factor of the fusion result;

[0119] D is the texture feature value of the image;

[0120] UA is the comprehensive contribution value of the initial gray value.

[0121] In this embodiment, the OP calculated by this submodule is a fusion calculation form based on OU and OW, which makes the three groups of algorithm units more correlated and logical. This algorithm unit can balance the grayscale information and color information in the final result, avoiding the limitations of a single information source, and can take into account a variety of image information to generate a more comprehensive and accurate comprehensive result. By adjusting the weight values ​​of g and h, the contribution of the initial grayscale image and the fusion result in OP can be controlled to achieve a balance of different information.

[0122] For example, when evaluating tooth alignment, the weight of the grayscale image in the comprehensive result can be increased;

[0123] When evaluating the jaw position, the weight of the fused image in the comprehensive result can be increased, and the image texture feature values ​​can be substituted to further improve the accuracy and richness of the comprehensive result. Therefore, this submodule can comprehensively consider a variety of image information to generate more comprehensive and accurate results, which is suitable for application scenarios with high real-time requirements;

[0124] This submodule can comprehensively evaluate the orthodontic effect. Combined with the grayscale submodule and the gray-color comprehensive submodule, it can comprehensively consider a variety of image information, and then conduct a comprehensive and accurate evaluation of the effect of orthodontic treatment, including the neatness of tooth arrangement, the correctness of jaw position, the relative relationship between teeth and jaw, etc. This submodule can also introduce additional information related to orthodontic treatment, such as the patient's age, gender, orthodontic treatment history, etc. This information can be used as an auxiliary factor to further improve the accuracy and reliability of orthodontic image detection.

[0125] See also Figures 1 to 4 , the processing process of adjusting the submodule is as follows:

[0126] According to the new red channel variable weight a n =a 1 +0.01*OP;

[0127] If a n >1, then output a n is 1, and a n Replace a in the grayscale submodule 1 ;

[0128] If a n <0, then output a n is 0, and a n Replace a in the grayscale submodule 1 ;

[0129] If 0 n ≤1, then a n Normal output, a n Replace a in the grayscale submodule​1 .

[0130] In this embodiment, by iteratively adjusting the variable weight of the red channel, the image detection system can dynamically optimize the brightness, contrast and detail performance of the grayscale image, especially in key areas such as teeth and jaws, so that the image is more suitable for analysis and evaluation of orthodontic treatment, and as the weights are adjusted by iteration, the overall system can gradually highlight the characteristics of the teeth and jaws, reduce the interference of background noise and irrelevant information, and thus improve the accuracy of orthodontic image detection. In addition, when applied to a variety of different orthodontic processes, this iterative form can adapt to a variety of orthodontic situations and provide doctors with more accurate and reliable diagnostic basis. Different from the existing fixed weights and difficult-to-adjust system methods, this system can achieve dynamic optimization and select appropriate weight values ​​according to specific image characteristics to improve image quality and detection accuracy. In the process of iterative optimization, it can also reduce the influence of interference factors such as image noise and illumination changes to a certain extent, improve the robustness and stability of the system, and can provide doctors with accurate evaluation of key information such as tooth arrangement and jaw position in orthodontic treatment, which is convenient for doctors' operation needs;

[0131] In the orthodontic images, the structural information of teeth and jaws is very important for diagnosis and treatment. By iteratively affecting the red channel, the red channel may have a higher sensitivity or contrast when displaying these structures. Therefore, adjusting the weight of the red channel can more effectively extract this information. 1 After iterative adjustments, the structural information of the teeth and jaws is more prominent in the image, and the contrast with the surrounding tissues is enhanced, making the details of these key areas more clearly discernible.

[0132] In the specific implementation process, the three submodules in the method are used to form an intelligent image detection system for oral orthodontic treatment. The red channel value, the green channel value and the blue channel value are substituted into the grayscale submodule to calculate the initial grayscale value. Different weights are assigned to the three color channels and the weighted grayscale value is calculated, thereby retaining part of the color information and reducing information loss, so as to improve the contrast and clarity of the grayscale image, make the boundary between the teeth and periodontal tissues clearer, and facilitate the subsequent feature extraction and detection;

[0133] By substituting the initial grayscale value and contrast value into the gray-color synthesis submodule, the image fusion value is calculated, and the grayscale image is fused with the original color image to generate an image that contains both grayscale information and retains part of the original color information. The fused image has both the clear boundary of the grayscale image and the color information of the color image, thereby providing a richer information source for subsequent image analysis and detection, and improving the accuracy of detection;

[0134] The initial grayscale value and the image fusion value are substituted into the comprehensive value of the image feature to calculate the comprehensive value of the image feature. This integrates a variety of image information and features to make the detection result more accurate and comprehensive. At the same time, the purpose of continuously optimizing multiple sub-modules in this method is achieved through iteration, so as to further improve the detection speed and efficiency and meet the real-time requirements of clinical applications. The three sub-modules can cooperate with each other in calculations, and can also perform overall cycles and iterations, so that the overall method has the effect of automatic optimization and updating, and thus better adaptability.

[0135] See also Figures 1 to 4 In this embodiment, the devices used by the output and application module are displays, printers, and treatment planning software, which output optimized images and calculation data, and apply the results to the formulation and evaluation of oral orthodontic treatment plans, thereby providing doctors with high-quality image data and analysis results, thereby assisting doctors in formulating treatment plans to improve patient treatment outcomes and patient satisfaction.

[0136] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent image detection method for orthodontic treatment, characterized in that: The image detection steps include the following: The first step is to scan the patient's oral cavity through the data acquisition module to obtain a three-dimensional point cloud model of the patient's jaws, and extract two-dimensional slice image data from the three-dimensional point cloud data; In the second step, the two-dimensional slice image data is preprocessed by the data processing module, and the initial weights of the red, green and blue channels are set; The third step is to convert the preprocessed two-dimensional slice image data into grayscale space, and obtain the initial grayscale value, image fusion value and image feature comprehensive value through the image processing module; The fourth step is to display the initial grayscale value, image fusion value and image feature comprehensive value on the user interface through the result output application module for the formulation and evaluation of the orthodontic treatment plan; The image processing module includes a grayscale submodule, a gray-color synthesis submodule, an image detection synthesis result submodule and an adjustment submodule; The processing process of the grayscale submodule is as follows: OU=(PA+PB+PC) 2 *B1+B2; PA=A1*a1, PB=A2*b, PC=A3*c; in: OU is the initial gray value; A1 is the red channel value, A2 is the green channel value, and A3 is the blue channel value; a1 is the variable weight of the red channel, b is the weight of the green channel, and c is the weight of the blue channel; B1 is the grayscale scaling factor value, and B2 is the grayscale offset value; PA is the red channel contribution value, PB is the green channel contribution value, and PC is the blue channel contribution value; The contribution values ​​of the three color channels, red channel value A1, green channel value A2, and blue channel value A3, are controlled by the red channel variable weight a1, the green channel weight b, and the blue channel weight c to highlight the features of the teeth and jaws in the grayscale image, and the range and offset of the grayscale value are adjusted by the grayscale value scaling factor value B1 and the grayscale value offset value B2 to optimize the brightness and contrast of the grayscale image, and an initial grayscale value OU is output; The processing process of the gray color comprehensive submodule is as follows: OW=d*OU+e*(A1+A2+A3)+PD; PD = f * C; in: OW is the image fusion value; d is the gray value weight factor, e is the weight factor of the sum of A1, A2 and A3; f is the contrast weight factor, C is the contrast value; PD is the contrast contribution value; The contrast of the fused image is enhanced by combining the contrast contribution value PD with the initial grayscale value OU output by the grayscale submodule, and the image fusion value OW is output; The processing process of the image detection comprehensive result submodule is as follows: OP = g*OU+OW*h+D; UA=g*OU in: OP is the comprehensive value of image features; g is the initial gray value weight factor, h is the fusion result weight factor, and D is the texture feature value of the image; UA is the comprehensive contribution value of the initial gray value; Based on the initial grayscale value OU output by the grayscale submodule and the image fusion value OW output by the gray-color synthesis submodule, the contribution of the initial grayscale value OU and the image fusion value OW in OP is controlled by adjusting the initial grayscale value weight factor g and the fusion result weight factor h to balance the initial grayscale value OU and the image fusion value OW, and output the image feature comprehensive value OP.

2. The intelligent image detection method for orthodontic treatment according to claim 1, characterized in that: The processing process of the adjustment submodule is as follows: According to the new red channel variable weight a n =a1+0.01*OP; If a n >1, then output a n is 1, and a n Replace a1 in the grayscale submodule; If a n <0, then output a n is 0, and a n Replace a1 in the grayscale submodule; If 0<a n ≤1, then a n Normal output, a n Replace a1 in the grayscale submodule; The image feature comprehensive value OP is input to the adjustment submodule, and the new red channel variable weight a is output. n , set the new red channel variable weight a n The original red channel variable weight a1 in the grayscale submodule is replaced to optimize the initial grayscale value OU, so that the initial grayscale value OU is continuously adjusted through iteration, and the number of iterations is set to six times. With the iteration, the original red channel variable weight a1 is adjusted to highlight the characteristics of teeth and jaws and improve the accuracy of orthodontic image detection.

3. The intelligent image detection method for orthodontic treatment according to claim 1, characterized in that: The devices used by the output and application module are a display, a printer and treatment planning software.

4. The intelligent image detection method for orthodontic treatment according to claim 1, characterized in that: The devices used by the data acquisition module are three-dimensional scanning equipment and image processing software.

5. The intelligent image detection method for orthodontic treatment according to claim 1, characterized in that: The data acquisition module includes an intraoral scanner.

Images