Palatal suture image reconstruction method and system based on Sobel operator

Abstract

The present invention relates to a method and system for reconstructing midpalatal suture images based on the Sobel operator, belonging to the technical field of image processing. Data preprocessing is first performed on a CBCT file of the maxillary midpalatal suture to obtain multiple local midpalatal suture images. These local midpalatal suture images are then merged and fused to obtain a reconstructed midpalatal suture image. During the merged fusion process, the Sobel operator is used for optimization, thereby reconstructing the full shape of the midpalatal suture region scattered across multi-layer CBCT images onto a single image, facilitating intuitive and accurate assessment of the midpalatal suture region by the physician.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a palatal suture image reconstruction method and system based on an optimized Sobel operator. Background Art

[0002] The midpalatal suture, located between the horizontal plates of the palatine bones, can be used to monitor maxillary width development and assess maxillary developmental potential in adolescents. However, the midpalatal suture exists as an arched area. Directly viewing the midpalatal suture in the CBCT file's tomographic images only shows a portion of the arched midpalatal suture, preventing doctors from visually perceiving or observing the entire suture. Consequently, it is difficult to accurately assess maxillary width development in adolescents based on the midpalatal suture's condition.

[0003] Based on this, there is an urgent need for a method and system that can observe the entire mid-palatal suture area in one image. Summary of the Invention

[0004] The purpose of the present invention is to provide a midpalatal suture image reconstruction method and system based on the Sobel operator, which can reconstruct the overall shape of the midpalatal suture area scattered on multi-layer CBCT images on the same image, making it easier for doctors to make intuitive and accurate judgments on the midpalatal suture area.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] A palatal suture image reconstruction method based on a Sobel operator, the reconstruction method comprising:

[0007] Data preprocessing was performed on the CBCT file of the maxillary palatal suture to obtain multiple local images of the palatal suture;

[0008] All the local images of the midpalatal suture are merged to obtain a reconstructed image of the midpalatal suture; wherein, in the process of merged fusion, the Sobel operator is used for optimization.

[0009] A palatal suture image reconstruction system based on a Sobel operator, the reconstruction system comprising:

[0010] A data preprocessing module is used to preprocess the data of the maxillary palatal suture CBCT file to obtain multiple local images of the palatal suture;

[0011] The fusion module is used to merge and fuse all the local images of the midpalatal suture to obtain a reconstructed image of the midpalatal suture; wherein, in the process of merge and fusion, the Sobel operator is used for optimization.

[0012] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0013] The present invention is used to provide a midpalatal suture image reconstruction method and system based on the Sobel operator. The method first preprocesses the data of the maxillary midpalatal suture CBCT file to obtain multiple midpalatal suture local images, and then merges and fuses all the midpalatal suture local images to obtain a midpalatal suture reconstructed image. In the process of merged fusion, the Sobel operator is used for optimization, so that the midpalatal suture area scattered on the multi-layer CBCT images can be reconstructed into its entire regional shape on the same image, which facilitates the doctor to intuitively and accurately judge the midpalatal suture area. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0015] Figure 1 A schematic diagram of the mid-palatal suture provided in Example 1 of the present invention;

[0016] Figure 2 A flowchart of the reconstruction method provided in Example 1 of the present invention;

[0017] Figure 3 This is a schematic diagram of the reconstruction method provided in Example 1 of the present invention;

[0018] Figure 4 This is a comparison chart of the effects of various reconstruction methods provided in Example 1 of the present invention;

[0019] Figure 5 This is a system block diagram of the reconstruction system provided by Example 2 of the present invention. DETAILED DESCRIPTION

[0020] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making creative efforts are within the scope of protection of the present invention.

[0021] The purpose of the present invention is to provide a method and system for reconstructing the midpalatal suture image based on the Sobel operator, which can extract, fuse and optimize the image of the maxillary midpalatal suture CBCT file, and reconstruct the overall shape of the midpalatal suture area scattered on multi-layer CBCT images on the same image, so as to facilitate the doctor to make an intuitive and accurate judgment on the midpalatal suture area.

[0022] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0023] Example 1:

[0024] like Figure 1 As shown, the palatal suture is an arched area, and it is impossible to show the full picture of the arched palatal suture on a single image. In order to show the full picture of the arched palatal suture on a single image as much as possible, the current method is: when taking medical CT images, take oblique shots at a certain angle according to the extension direction of the palatal suture, thereby obtaining a series of oblique CBCT images, and then find an oblique image that is closest to the palatal suture angle and contains the largest palatal suture area, so as to restore the full picture of the palatal suture area to the greatest extent. However, this method will bring two problems: (1) It is difficult to find the shooting angle: Due to the differences in the human body, the palatal suture area of ​​each person has different arches and tilts. Therefore, it is difficult to confirm at which angle to take the oblique image before taking the oblique image in order to restore the full picture of the palatal suture area to the greatest extent. (2) It is difficult to restore the full picture: Since the maxillary palatal suture area is an arched area, for people with a small arch, the oblique image can indeed show the full picture of the palatal suture to a certain extent. However, for people with a large palatal suture arch, the oblique image can still only reflect part of the palatal suture area morphology, and cannot show the full picture of the palatal suture area. Overall, this method has disadvantages such as low practicality and poor versatility, which means it is difficult to promote and use on a large scale.

[0025] With the development of artificial intelligence, the palatal suture reconstruction technology based on Sobel operator technology has become possible and necessary. The method of fusing palatal suture area images based on conventional CBCT images has the characteristics of easy implementation and strong versatility, and has good application prospects. Based on this, this embodiment combines the knowledge of domain experts to design a set of automated, intelligent, and high-precision palatal suture image reconstruction technology, and provides a palatal suture image reconstruction method based on Sobel operator, such as Figure 2 and Figure 3 As shown, the reconstruction method includes:

[0026] S1: Data preprocessing of the maxillary palatal suture CBCT file to obtain multiple local images of the palatal suture;

[0027] The CBCT file of the maxillary palatal suture can be directly retrieved from the hospital database. The data preprocessing steps of the maxillary palatal suture CBCT file, i.e., S1, may include:

[0028] (1) Reading the CBCT file of the maxillary palatal suture to obtain file information, including the resolution, number of slices, window width, and window position of the CBCT file of the maxillary palatal suture;

[0029] (2) Convert the maxillary palatal suture CBCT file into a three-dimensional grayscale matrix based on the file information, and then convert the three-dimensional grayscale matrix into multiple axial CT cross-sectional images;

[0030] The resolution and number of slices determine the three-dimensional shape and size of the three-dimensional grayscale matrix, while the window width and window level determine the grayscale interval of the three-dimensional grayscale matrix. In this embodiment, converting a CBCT file of the maxillary palatal suture into a three-dimensional grayscale matrix based on file information may include: determining the three-dimensional size of the three-dimensional grayscale matrix based on the resolution and number of slices, determining the grayscale interval of the three-dimensional grayscale matrix based on the window width and window level, and then converting the CBCT file of the maxillary palatal suture into a three-dimensional grayscale matrix based on the three-dimensional size and grayscale interval. For example, the resolution and number of slices determine the structure of the three-dimensional grayscale matrix. For example, if the resolution is 512*512 and the number of slices is 400, the three-dimensional size of the three-dimensional grayscale matrix is ​​512*512*400. Window width and window level are terms used in medical CT. A simple calculation of the window width and window level will result in the grayscale interval of the three-dimensional grayscale matrix. This grayscale interval determines which tissue (such as blood vessels, organs, or bones) is more prominent in the image.

[0031] Converting the three-dimensional grayscale matrix into multiple axial CT cross-sectional images may include: segmenting the three-dimensional grayscale matrix in the height direction, converting the three-dimensional grayscale matrix into a series of axial CT cross-sectional images, and uniformly converting the axial CT cross-sectional images into a 512*512 resolution. Axial CT cross-sectional images refer to images parallel to a plane defined by the x and y axes.

[0032] (3) Each axial CT cross-sectional image containing the midpalatal suture area is cut to obtain multiple local images of the midpalatal suture.

[0033] Image recognition was performed on all axial CT cross-sections, and the axial CT cross-sections containing the midpalatal suture region were extracted. The midpalatal suture region in each axial CT cross-section containing the midpalatal suture region was then cut to obtain multiple local images of the midpalatal suture. After cutting, all local images of the midpalatal suture were normalized and uniformly converted to a 50*200 resolution.

[0034] S2: performing a merge-type fusion on all the local images of the midpalatal suture to obtain a reconstructed image of the midpalatal suture; wherein, during the merge-type fusion process, a Sobel operator is used for optimization.

[0035] The convolution operator-based image fusion method used in this embodiment is the product of communication with experts and optimization of existing pixel-level image fusion algorithms. It mainly adjusts the fusion weights by combining the existing pixel-level image fusion algorithm with the characteristics of the midpalatal suture area, and uses the Sobel operator to optimize the fused image, ultimately obtaining a high-definition and textured midpalatal suture reconstructed image. The image fusion processing step, i.e., S2, may include:

[0036] (1) All the local images of the midpalatal suture are combined in pairs to obtain multiple groups of image pairs to be fused; the pixels of the local images of the midpalatal suture included in each group of image pairs to be fused correspond to each other, and the corresponding pixels are recorded as pixel pairs;

[0037] All midpalatal suture images are paired together by randomly but non-repeatedly selecting two midpalatal suture images as a pair of images to be fused. If the number of midpalatal suture images is odd, the remaining midpalatal suture image is used as a single pair of images to be fused. For example, if there are five midpalatal suture images, the first and second midpalatal suture images can be used as a pair of images to be fused, the third and fifth midpalatal suture images as a pair of images to be fused, and the fourth midpalatal suture image as a pair of images to be fused, for a total of three pairs of images to be fused. When an image pair to be fused includes two midpalatal suture images, the pixels of the two images are mapped one-to-one, and the corresponding pixels are recorded as a pixel pair. When an image pair to be fused includes only one midpalatal suture image, the pixels of the one image are recorded as a pixel pair.

[0038] Assuming that the number of local images of the midpalatal suture is N, after combining all the local images of the midpalatal suture in pairs, the number of image pairs to be fused is: in, Represents the operator symbol for rounding down. If N=4, then after pairwise combination, we have If N=5, then after pairwise combination, we have

[0039] (2) For each set of image pairs to be fused, the fusion weight of each pixel pair is calculated, and the image pairs to be fused are merged and fused according to the fusion weight of each pixel pair to obtain the fused image;

[0040] Among them, calculating the fusion weight of each pixel pair can include: calculating the total average grayscale and adjustment factor based on the grayscale values ​​of all local images of the midpalatal suture, specifically, calculating the average grayscale value of all pixel points of all local images of the midpalatal suture to obtain the total average grayscale, calculating the grayscale difference of the grayscale values ​​of any two pixel points of all local images of the midpalatal suture to determine the maximum grayscale difference of all local images of the midpalatal suture, and determining the adjustment factor based on the maximum grayscale difference; for each pixel pair, calculating the average grayscale value of the image pair to be fused in the pixel pair to obtain the average grayscale; calculating the fusion weight of the pixel pair based on the total average grayscale, the adjustment factor and the average grayscale.

[0041] The calculation formula of the pixel pair fusion weight is:

[0042]

[0043] In formula (1), M ij is the fusion weight of pixel pair ij (i.e., the pixel pair consisting of pixel i of one local image of the midpalatal suture and pixel j of another local image of the midpalatal suture, and the positions of pixel i and pixel j correspond to each other); A ij is the average grayscale of pixel pair ij; e is the total average grayscale; d is the adjustment factor. It should be noted that if the pixel pair to be fused only includes one local image of the palatal suture, the pixel pair ij is a single pixel point, and the grayscale value of the pixel point is directly used as the average grayscale A. ij .

[0044] Merging and fusing the image pairs to be fused according to the fusion weight of each pixel pair to obtain the fused image may include: for each pixel pair, calculating the average grayscale value of the image pair to be fused in the pixel pair to obtain the average grayscale; calculating the fused grayscale value of the pixel pair according to the average grayscale and the fusion weight to obtain the fused image.

[0045] The calculation formula of the gray value after fusion is:

[0046]

[0047] In formula (2), P ij is the gray value of the fused pixel pair ij.

[0048] After obtaining the fused grayscale values ​​of all pixel pairs, the positions of all pixel pairs are kept unchanged, and the fused grayscale values ​​are used as the grayscale values ​​of the pixels at that position to obtain the fused image.

[0049] In this embodiment, the fusion weight of each pixel pair is calculated according to the grayscale value of the pixel pair. Based on the calculated fusion weight, the N cut local images of the midpalatal suture are merged and fused in pairs to obtain a fused image.

[0050] (3) Optimize each fused image using the Sobel operator to obtain the optimized image;

[0051] The fused image is optimized using the Sobel operator, making the high-frequency regions of the resulting optimized image more pronounced and the overall contrast higher. This avoids issues such as image fogging and blurred textures typically associated with image fusion. Specifically, for each pixel in the fused image, the optimized grayscale value is calculated using the Sobel operator to produce the optimized image. After obtaining the optimized grayscale values ​​for all pixels, the fused grayscale values ​​are replaced with the optimized grayscale values, maintaining their positions unchanged, to produce the optimized image.

[0052] (4) Determine whether the number of optimized images is 1;

[0053] (5) If yes, the optimized image is used as the reconstructed image of the midpalatal suture;

[0054] (6) If not, the optimized image is used as the local image of the midpalatal suture in the next cycle, and the process returns to the step of "combining all the local images of the midpalatal suture in pairs".

[0055] The optimized images are merged and fused again, i.e., steps (1) to (3) are repeated until all local images of the midpalatal suture are fused into a midpalatal suture overall image, and a midpalatal suture reconstructed image is obtained.

[0056] Here, we introduce the optimization principle and implementation process of the Sobel operator:

[0057] The Sobel operator (Sobel operator) is mainly used for edge detection. It is a discrete difference operator used to calculate the grayscale approximation of the image brightness function. Using this operator at any pixel in the image will generate a corresponding grayscale vector or normal vector. The operator contains 8 different types of 3*3 matrices, namely up, down, left, right, upper left, lower left, upper right, and lower right. By performing a planar convolution on the operator and the image, the horizontal and vertical brightness difference approximations can be obtained respectively. If A represents the original image, G x and G y Represent the grayscale values ​​of the image after horizontal and vertical edge detection respectively. The formula for planar convolution of the Sobel operator and the image is as follows:

[0058]

[0059] The two matrices in formula (3) are Sobel operators.

[0060] The specific calculation formula is as follows:

[0061] Gx =(-1)*f(x-1,y-1)+0*f(x,y-1)+1*f(x+1,y-1)+(-2)*f(x-1,y)+0*f(x,y)+2*f(x+1,y)+(-1)*f(x-1,y+1)+0 *f(x,y+1)+1*f(x+1,y+1)=[f(x+1,y-1)+2*f(x+1,y)+f(x+1,y+1)]-[f(x-1,y-1)+2*f(x-1,y)+f(x-1,y+1)]

[0062] G y =1*f(x-1,y-1)+2*f(x,y-1)+1*f(x+1,y-1)+0*f(x-1,y)+0*f(x,y)+0*f(x+1,y)+(-1)*f(x-1,y+1)+(-2)*f (x,y+1)+(-1)*f(x+1,y+1)=[f(x-1,y-1)+2f(x,y-1)+f(x+1,y-1)]-[f(x-1,y+1)+2*f(x,y+1)+f(x+1,y+1)]

[0063] Where f(a, b) represents the grayscale value of the pixel in image (a, b).

[0064] The grayscale value of each pixel (x, y) in the image after horizontal and vertical edge detection is combined with the following formula to calculate the grayscale value of the pixel (x, y):

[0065]

[0066] If the gray value G is greater than a certain threshold, the point (x, y) is considered to be an edge point.

[0067] In terms of image optimization, this embodiment directly uses the Sobel operator with appropriate weights to convolve the image, which can sharpen the image edges to a certain extent and enhance the texture. Specifically, in image processing, there is a basic processing method: linear filtering. The planar digital image to be processed can be regarded as a large matrix. Each pixel of the planar digital image to be processed corresponds to each element of the large matrix. A small filter matrix (i.e., a convolution operator) is used for filtering. The small filter matrix is ​​generally a square matrix, that is, the number of rows and columns is the same. Filtering is to calculate the product of the surrounding pixels and the corresponding position elements of the small filter matrix for each pixel in the large matrix, and then add the results together. The final value is used as the new value of the pixel, thus completing a filtering. In this embodiment, the fused image is regarded as a large matrix. Each pixel of the fused image corresponds to each element of the large matrix. The Sobel operator is regarded as a small filter matrix used for filtering. For each pixel in the fused image, the optimized grayscale value G of the pixel is calculated using equations (3) and (4). The positions of all pixels are kept unchanged, and the fused grayscale value is replaced by the optimized grayscale value G to obtain the optimized image.

[0068] As an optional implementation, after completing a fusion, that is, after a complete merging to obtain the palatal suture reconstructed image, the texture features of the palatal suture reconstructed image are analyzed, and according to the effect of the palatal suture reconstructed image and combined with expert opinions, the weight of the Sobel operator is adjusted, that is, the element value of the matrix in formula (3) is adjusted, and the image fusion optimization is performed again. After multiple fusion optimizations, the optimal Sobel operator is selected according to the optimization situation, which can achieve the best fusion effect, and finally solve the problems of blurred boundaries and unclear textures caused by fusion, and achieve the purpose of image optimization. Figure 4 As shown, it is a schematic diagram of the fusion effect obtained by various fusion methods. It can be seen that the fusion effect of the fusion method of this embodiment is very excellent.

[0069] Specifically, after obtaining the reconstructed image of the midpalatal suture, the reconstruction method of this embodiment further includes:

[0070] (1) Analyze the texture features of the reconstructed image of the midpalatal suture to adjust the weight of the Sobel operator and obtain the optimized Sobel operator;

[0071] (2) All local images of the midpalatal suture are merged and fused to obtain a new midpalatal suture reconstructed image; wherein, in the process of merged fusion, the optimized Sobel operator is used for optimization;

[0072] (3) Determine whether the maximum number of iterations has been reached;

[0073] (4) If yes, then determine the optimal Sobel operator based on the optimization of all reconstructed images; the reconstructed images include the midpalatal suture reconstructed image and the new midpalatal suture reconstructed image;

[0074] (5) If not, the new palatal suture reconstructed image is used as the palatal suture reconstructed image in the next cycle, and the optimized Sobel operator is used as the Sobel operator in the next cycle, and the process returns to the step of "analyzing the texture features of the palatal suture reconstructed image".

[0075] The palatal suture reconstruction method of this embodiment is mainly divided into three parts, namely CBCT file preprocessing, image fusion based on the Sobel operator, and analysis of the palatal suture processing results. After obtaining the CBCT file, data preprocessing is first performed to process the CBCT file into a three-dimensional grayscale matrix, and then into an axial layered image, and the palatal suture area is cut to obtain a local image of the palatal suture. Finally, the cut local image of the palatal suture is normalized. After completing the data preprocessing, the local images of the palatal suture are merged and fused. The fusion parameters and weights are calculated before each fusion. After each round of fusion is completed, the fused image is optimized using the Sobel operator to finally obtain a fused image of all the local images of the palatal suture, which is approximately a projection of the entire morphology of the palatal suture on the axial plane. This reconstruction method can be used to detect the development of the maxillary width and evaluate the maxillary development potential of adolescents by extracting, fusing and optimizing the CBCT files of the maxillary palatal suture. This reconstruction method can bring the following benefits:

[0076] (1) It is difficult and too subjective for the human eye to directly analyze the overall picture of the midpalatal suture from the multi-slice CBCT file. It is impossible to intuitively feel the morphology of the midpalatal suture area and make a judgment on the development of the maxillary bone of adolescents based on it. However, the reconstruction method of this embodiment can directly observe the overall projection of the midpalatal suture morphology in one image.

[0077] (2) This embodiment can directly reconstruct the CT image of the midpalatal suture region using conventional CBCT files, without the need for additional CBCT examinations at specific angles, greatly reducing the difficulty of popularization and widespread use. The process of reconstructing the midpalatal suture image is minimally affected by the arch of the midpalatal suture region of the patient. It is applicable to patients with large or small arches of the midpalatal suture, making it more versatile than other methods.

[0078] Compared to the traditional oblique CBCT method of acquiring the midpalatal suture morphology, the midpalatal suture reconstruction method based on Sobel operator technology adopted in this embodiment directly fuses and reconstructs the midpalatal suture region contained in conventional CBCT multi-slice images, restoring the full morphological picture of the midpalatal suture region as a plane projection. This process does not require any shooting angle requirements and can also display the full morphological picture of the midpalatal suture in a single image for patients with large midpalatal suture arches. The application of this embodiment saves time and effort for doctors and allows for a more intuitive acquisition of the full morphological picture of the midpalatal suture.

[0079] Example 2:

[0080] This embodiment is used to provide a palatal suture image reconstruction system based on the Sobel operator. Figure 5 As shown, the reconstruction system includes:

[0081] The data preprocessing module M1 is used to preprocess the data of the maxillary palatal suture CBCT file to obtain multiple local images of the palatal suture;

[0082] The fusion module M2 is used to perform a merge-type fusion on all the local images of the midpalatal suture to obtain a reconstructed image of the midpalatal suture; wherein, in the merge-type fusion process, the Sobel operator is used for optimization.

[0083] The fusion module may include:

[0084] a combining unit, configured to combine all the local images of the midpalatal suture in pairs to obtain a plurality of image pairs to be fused; wherein the pixels of the local images of the midpalatal suture included in each group of the image pairs to be fused correspond to each other, and the corresponding pixels are recorded as a pixel pair;

[0085] a fusion unit, configured to calculate, for each set of the image pairs to be fused, a fusion weight of each pixel pair, and merge and fuse the image pairs to be fused according to the fusion weight of each pixel pair to obtain a fused image;

[0086] an optimization unit, configured to optimize each of the fused images using a Sobel operator to obtain an optimized image;

[0087] A judging unit, configured to judge whether the number of the optimized images is 1;

[0088] a reconstruction unit, configured to use the optimized image as a reconstructed image of the midpalatal suture if the optimized image is true;

[0089] The return unit is used to use the optimized image as the local image of the midpalatal suture in the next cycle if not, and return to the step of "combining all the local images of the midpalatal suture in pairs".

[0090] Each embodiment in this specification focuses on the differences from other embodiments, and the same or similar parts between the embodiments can be referred to each other. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0091] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The above examples are only intended to help understand the method and core concept of the present invention. At the same time, those skilled in the art will find that the specific implementation methods and application scopes may vary based on the concept of the present invention. In summary, the contents of this specification should not be construed as limiting the present invention.

Claims

1. A palatal suture image reconstruction method based on Sobel operator, characterized in that: The reconstruction method includes: Data preprocessing was performed on the CBCT file of the maxillary palatal suture to obtain multiple local images of the palatal suture; Performing a merge-type fusion on all the local images of the midpalatal suture to obtain a reconstructed image of the midpalatal suture; wherein, in the process of the merge-type fusion, a Sobel operator is used for optimization; The merging and fusing of all the local images of the midpalatal suture to obtain a reconstructed image of the midpalatal suture specifically includes: All the local images of the midpalatal suture are combined in pairs to obtain multiple groups of image pairs to be fused; the pixels of the local images of the midpalatal suture included in each group of the image pairs to be fused correspond to each other, and the corresponding pixels are recorded as a pixel pair; For each group of the image pairs to be fused, calculating the fusion weight of each pixel pair, and merging and fusing the image pairs to be fused according to the fusion weight of each pixel pair to obtain a fused image; Optimizing each of the fused images using a Sobel operator to obtain an optimized image; Determine whether the number of the optimized images is 1; If yes, the optimized image is used as the reconstructed image of the midpalatal suture; If not, the optimized image is used as the local image of the midpalatal suture in the next cycle, and the process returns to the step of "combining all the local images of the midpalatal suture in pairs".

2. The reconstruction method according to claim 1, characterized in that The data preprocessing of the CBCT file of the maxillary palatal suture to obtain multiple local images of the palatal suture specifically includes: Reading a CBCT file of the maxillary palatal suture to obtain file information; the file information includes a resolution, number of slices, window width, and window position of the CBCT file of the maxillary palatal suture; Converting the maxillary palatal suture CBCT file into a three-dimensional grayscale matrix according to the file information, and converting the three-dimensional grayscale matrix into multiple axial CT cross-sectional images; Each of the axial CT transverse images containing the midpalatal suture area is cut to obtain multiple local images of the midpalatal suture.

3. The reconstruction method according to claim 2, characterized in that The converting of the maxillary palatal suture CBCT file into a three-dimensional grayscale matrix according to the file information specifically includes: determining a three-dimensional size of a three-dimensional grayscale matrix according to the resolution and the number of layers; Determining the grayscale interval of the three-dimensional grayscale matrix according to the window width and the window position; The maxillary palatal suture CBCT file is converted into a three-dimensional grayscale matrix according to the three-dimensional size and the grayscale interval.

4. The reconstruction method according to claim 1, characterized in that The calculating of the fusion weight of each pixel pair specifically includes: Calculating the total average grayscale and adjustment factor according to the grayscale values ​​of all the local images of the midpalatal suture; For each pixel pair, calculating the average grayscale value of the image pair to be fused in the pixel pair to obtain an average grayscale; The fusion weight of the pixel pair is calculated according to the total average grayscale, the adjustment factor and the average grayscale.

5. The reconstruction method according to claim 1, characterized in that: Merging and fusing the image pairs to be fused according to the fusion weight of each pixel pair to obtain a fused image specifically includes: For each pixel pair, calculating the average grayscale value of the image pair to be fused in the pixel pair to obtain an average grayscale; The fused grayscale value of the pixel pair is calculated according to the average grayscale and the fusion weight to obtain a fused image.

6. The reconstruction method according to claim 1, characterized in that Optimizing each of the fused images by using the Sobel operator to obtain the optimized image specifically includes: For each pixel of each fused image, the optimized grayscale value of the pixel is calculated based on the Sobel operator to obtain the optimized image.

7. The reconstruction method according to claim 1, characterized in that: After obtaining the reconstructed image of the midpalatal suture, the reconstruction method further includes: Analyzing the texture features of the reconstructed image of the midpalatal suture to adjust the weight of the Sobel operator to obtain an optimized Sobel operator; Performing a merge-type fusion on all the local images of the midpalatal suture to obtain a new midpalatal suture reconstructed image; wherein, in the process of the merge-type fusion, the optimized Sobel operator is used for optimization; Determine whether the maximum number of iterations has been reached; If yes, then determining the optimal Sobel operator according to the optimization conditions of all reconstructed images; the reconstructed images include the midpalatal suture reconstructed image and the new midpalatal suture reconstructed image; If not, the new palatal suture reconstructed image is used as the palatal suture reconstructed image in the next cycle, and the optimized Sobel operator is used as the Sobel operator in the next cycle, and the process returns to the step of "analyzing the texture features of the palatal suture reconstructed image".

8. A palatal suture image reconstruction system based on Sobel operator, characterized in that: The reconstruction system comprises: A data preprocessing module is used to preprocess the data of the maxillary palatal suture CBCT file to obtain multiple local images of the palatal suture; A fusion module is used to perform a merge-type fusion on all the local images of the midpalatal suture to obtain a reconstructed image of the midpalatal suture; wherein, during the merge-type fusion process, a Sobel operator is used for optimization; The fusion module specifically includes: a combining unit, configured to combine all the local images of the midpalatal suture in pairs to obtain a plurality of image pairs to be fused; wherein the pixels of the local images of the midpalatal suture included in each group of the image pairs to be fused correspond to each other, and the corresponding pixels are recorded as a pixel pair; a fusion unit, configured to calculate, for each set of the image pairs to be fused, a fusion weight of each pixel pair, and merge and fuse the image pairs to be fused according to the fusion weight of each pixel pair to obtain a fused image; an optimization unit, configured to optimize each of the fused images using a Sobel operator to obtain an optimized image; A judging unit, configured to judge whether the number of the optimized images is 1; a reconstruction unit, configured to use the optimized image as a reconstructed image of the midpalatal suture if the optimized image is true; The return unit is used to, if not, use the optimized image as the local image of the midpalatal suture in the next cycle and return to the step of "combining all the local images of the midpalatal suture in pairs".

Images