Automatic identification method of microfractures in bone and joint X-rays based on edge detection

Abstract

The present invention relates to the technical field of bone joint X-ray image segmentation, and specifically to a method for automatically identifying microfractures in bone joint X-rays based on edge detection. The present invention obtains the bone joint surface edge coefficient based on the grayscale distribution and gradient distribution of pixel points in the bone joint X-ray image, and then obtains the bone joint area; obtains the weak edge coefficient of the edge area based on the distribution of edge pixel points in the area; obtains the number of fused pixel points of the outer pixel points based on the grayscale difference between the outer pixel points and the outer adjacent pixel points in the edge area; obtains the fusion coefficient of the edge area in combination with the weak edge coefficient; and locates microfractures based on the fusion coefficient of the edge area in the bone joint stratification. The present invention can identify the fracture line caused by microfractures, so that relevant personnel can accurately identify the microfracture area appearing in the patient's body.

Description

Technical Field

[0001] The invention relates to the technical field of bone and joint X-ray image segmentation, and in particular to an automatic identification method for bone and joint X-ray microfractures based on edge detection. Background Art

[0002] Microfractures often occur in complex joints, lumbar spine and other parts of the body. The fracture site is usually hidden, the range of fracture damage is relatively small, and the early symptoms are mild. Currently, digital technology is often used in clinical diagnosis of microfractures. Line photography, through the acquisition of The diagnostic results can be obtained by imaging diagnosis with optical images. However, due to the characteristics of microfractures, in the specific clinical application process, Optical images may lead to a certain degree of missed diagnosis or misdiagnosis, which may have adverse effects on subsequent treatment, so it is necessary to improve The correct diagnosis rate of microfractures in optical images can be improved, and the occurrence of missed diagnoses and misdiagnoses can be reduced, which can reduce the adverse effects on the patient's subsequent treatment. Image processing technology can be used to identify microfractures with higher accuracy, reduce the intervention of human subjective factors, ensure the consistency of diagnostic results, and improve Correct diagnosis rate of microfractures in optical images.

[0003] Edge detection, as a method of image detection of target edges, is more targeted at detecting fracture lines in X-ray images. However, in the prior art, the fracture lines of micro-fractures of bone joints are relatively subtle, resulting in unclear fracture lines. Traditional edge detection algorithms are unable to accurately identify fracture lines, resulting in the inability of relevant personnel to accurately identify fracture lines caused by micro-fractures in the patient's body. Summary of the invention

[0004] In order to solve the technical problem in the prior art that the fracture line of micro-fractures of bone and joints is relatively fine, resulting in unclear fracture lines, and traditional edge detection algorithms are unable to accurately identify fracture lines, thereby resulting in the inability of relevant personnel to accurately identify fracture lines caused by micro-fractures occurring in the patient's body, the purpose of the present invention is to provide an automatic identification method for micro-fractures in bone and joint X-rays based on edge detection, and the adopted technical scheme is specifically as follows: an automatic identification method for micro-fractures in bone and joint X-rays based on edge detection, the method comprising: acquiring a bone and joint X-ray image of a patient; obtaining a bone and joint surface edge coefficient of each pixel point in the bone and joint X-ray image according to the grayscale distribution and gradient distribution of the pixel points in the bone and joint X-ray image; obtaining a bone and joint area according to the bone and joint surface edge coefficient; obtaining all edge areas in the bone and joint area; and obtaining all edge areas in the bone and joint area according to the bone and joint area. According to the position distribution and gradient distribution of the middle edge pixel points, the weak edge coefficient of each edge area in the bone joint area is obtained; according to the position distribution of the pixel points in the bone joint area, the outer edge pixel points in each edge area are obtained; all the outer adjacent pixel points of the outer edge pixel points in each edge area are obtained; according to the grayscale difference between each outer edge pixel point in each edge area and each outer adjacent pixel point in the preset neighborhood, the number of fused pixel points in the preset neighborhood of each outer edge pixel point in each edge area is obtained; according to the number of fused pixel points and the weak edge coefficient of the outer edge pixel point, the fusion coefficient of each edge area is obtained; the bone joint area is stratified to obtain each bone joint stratification; the area where the microfracture is located is located according to the fusion coefficient of all edge areas in each bone joint stratification.

[0005] Furthermore, the method for obtaining the bone joint surface edge coefficient includes: obtaining the bone joint surface edge coefficient according to a bone joint surface edge coefficient calculation formula, and the bone joint surface edge coefficient calculation formula is as follows: In the formula, Indicates the first The edge coefficient of the bone joint surface at each pixel; Indicates the first The gradient value of each pixel; Indicates the number of other pixels in the preset neighborhood of each pixel in the bone joint X-ray image; Indicates the first The gray value of each pixel; Indicates the first The preset neighborhood of pixels The gray value of each pixel; Indicates the first The standard deviation of the gray value difference between a pixel and all pixels in the preset neighborhood; Represents an exponential function with a natural constant as base.

[0006] Furthermore, the method for acquiring the bone joint area includes: taking pixel points whose bone joint surface edge coefficient is greater than a preset first threshold as bone joint surface pixel points; and taking the area composed of all the bone joint surface pixel points as the bone joint area.

[0007] Furthermore, the method for obtaining the weak edge coefficient includes: obtaining all edge regions composed of all edge pixels in the bone joint region according to an edge detection algorithm; obtaining the weak edge coefficient according to a weak edge coefficient calculation formula, and the weak edge coefficient calculation formula is as follows: In the formula, Indicates the weak edge coefficient of the edge region in the bone joint region; Indicates the maximum number of pixels with the same and continuous gradient direction in the edge area; Indicates the number of edge pixels in the edge area; Represents the mean gradient value of other pixels in the edge area except for the pixels with the same and continuous gradient direction; Represents the horizontal coordinate of the edge starting point in the edge area; The ordinate of the edge starting point in the edge region; The horizontal coordinate of the edge end point in the edge area; The ordinate of the edge end point in the edge region; Represents an exponential function with a natural constant as base.

[0008] Furthermore, the method for acquiring the outer edge pixel points includes: in the edge area, taking corresponding edge pixel points whose number of pixel points other than the edge pixel points in a preset neighborhood is not 0 as outer edge pixel points.

[0009] Furthermore, the method for obtaining the number of fused pixel points includes: obtaining the number of fused pixel points according to a calculation formula for the number of fused pixel points, and the calculation formula for the number of fused pixel points is as follows: In the formula, Indicates the edge area The number of fused pixels within the preset neighborhood of the outer edge pixels; Indicates the edge area Gray value of the outer edge pixel; Indicates the edge area The first pixel in the preset neighborhood of the outer edge pixel The gray value of each pixel; Indicates the number of edge pixels in the edge area; Indicates the edge area Gray value of edge pixels; represents the sum function; Represents an Iverson bracket, where when the condition inside the bracket is met, the value of the bracket is 1, and when the condition inside the bracket is not met, the value of the bracket is 0.

[0010] Furthermore, the method for obtaining the fusion coefficient includes: accumulating and normalizing the number of fused pixels of each outer edge pixel to obtain a fusion weight of the edge area; and taking the product of the weak edge coefficient of the edge area and the fusion weight as the fusion coefficient of the edge area.

[0011] Furthermore, the method for acquiring bone joint stratification includes: stratifying the bone joint area according to a preset stratification threshold to obtain all bone joint stratifications.

[0012] Furthermore, the area where the microfracture is located is located according to the fusion coefficient of the edge area in each bone joint layer, including: normalizing the product of the fusion coefficient of the edge area and the number of bone joint layer layers passed by the edge area as the microfracture location score of the edge area; and taking the edge area with a microfracture location score greater than a preset second threshold as the area where the microfracture is located.

[0013] A system for automatically identifying microfractures in bone and joint X-rays based on edge detection, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above method when executing the computer program.

[0014] The present invention has the following beneficial effects: the present invention obtains bone joint X-ray images; since there is a certain adhesion between the bone joint area and other tissues, in order to study the micro-fracture area in the bone joint, the bone joint area is firstly obtained; since there is an obvious boundary between the bone joint area and other tissues in the bone joint X-ray image, the boundary of the bone joint area is firstly located, and since the edge pixel points of the bone joint area boundary have a large difference in grayscale and gradient from the pixel points of other tissues, the bone joint surface edge coefficient of each pixel point in the bone joint X-ray image is obtained according to the grayscale distribution and gradient distribution of the pixel points in the bone joint X-ray image; since the gradient directions of the edge pixel points of the weak edge are not consistent and the pixel points are not clear, the edge image in the bone joint area is obtained according to the grayscale distribution and gradient distribution of the pixel points in the bone joint X-ray image. The position distribution and gradient distribution of the pixel points are used to obtain the weak edge coefficient of the edge area in the bone joint area; since the edge of a microfracture is not easy to distinguish from the adjacent edge area, the embodiment of the present invention fuses the outer edge pixel points of the edge area with the outer adjacent pixel points to enhance the edge area, so the fusion coefficient of the edge area is analyzed by the number of pixels that can be fused in the edge area to obtain the fusion coefficient of the edge area; in order to avoid the possible small difference in fusion coefficients of different edge areas, which ultimately makes it impossible to accurately locate the microfracture area, the bone joint area is stratified to obtain each bone joint stratification; the area where the microfracture is located is located according to the fusion coefficients of all edge areas in each bone joint stratification. The present invention can identify the fracture line caused by microfractures, so that relevant personnel can accurately identify the microfracture area in the patient's body. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. 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 creative work.

[0016] Figure 1 A flow chart of a method for automatically identifying microfractures in bone and joint X-rays based on edge detection provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the following is a detailed description of the specific implementation method, structure, characteristics and effects of an automatic identification method of microfractures in bone and joint X-rays based on edge detection proposed by the present invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0018] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0019] The following is a detailed description of a specific scheme of the method for automatically identifying microfractures in bone and joint X-rays based on edge detection provided by the present invention in conjunction with the accompanying drawings.

[0020] See also Figure 1 , which shows a method for automatically identifying microfractures in bone and joint X-rays based on edge detection provided by an embodiment of the present invention. The method includes: Step S1: Acquire a bone and joint X-ray image of a patient.

[0021] The embodiment of the present invention is mainly applied to the positioning scene of the bone joint microfracture area, so the embodiment of the present invention first obtains the bone joint image. Since the bone joints of the patient are often observed by X-ray technology in the prior art, the bone joint X-ray image of the patient is obtained.

[0022] Grayscaling can reduce the amount of image data because each pixel only needs one grayscale value to represent it instead of three color channels, thereby simplifying the image processing algorithm and reducing the computational complexity and required storage space. Grayscaling helps to highlight important features in the image, such as edge features, while denoising can reduce random noise in the image, making the structure and features in the image clearer. Therefore, in an embodiment of the present invention, grayscaling and denoising are performed on the bone joint X-ray image for image preprocessing to obtain the final bone joint X-ray image.

[0023] Step S2: According to the grayscale distribution and gradient distribution of the pixels in the bone joint X-ray image, obtain the bone joint surface edge coefficient of each pixel in the bone joint X-ray image; obtain the bone joint area according to the bone joint surface edge coefficient; obtain all edge areas in the bone joint area; according to the position distribution and gradient distribution of the edge pixels in the bone joint area, obtain the weak edge coefficient of each edge area in the bone joint area.

[0024] In reality, there is a certain adhesion between the bone joint area and other tissues. In order to study the micro-fracture area in the bone joint, the bone joint area is first acquired. Since there is an obvious boundary between the bone joint area and other tissues in the bone joint X-ray image, in an embodiment of the present invention, the boundary of the bone joint area is first located. Since the grayscale and gradient of the edge pixels of the bone joint area boundary are greatly different from those of the pixels of other tissues, the bone joint surface edge coefficient of each pixel in the bone joint X-ray image is obtained based on the grayscale distribution and gradient distribution of the pixels in the bone joint X-ray image.

[0025] Preferably, in one embodiment of the present invention, the method for obtaining the bone joint surface edge coefficient includes: obtaining the bone joint surface edge coefficient according to a bone joint surface edge coefficient calculation formula, and the bone joint surface edge coefficient calculation formula is as follows: In the formula, Indicates the first The edge coefficient of the bone joint surface at each pixel; Indicates the first The gradient value of each pixel; Indicates the number of other pixels in the preset neighborhood of each pixel in the bone joint X-ray image; Indicates the first The gray value of each pixel; Indicates the first The preset neighborhood of pixels The gray value of each pixel; Indicates the first The standard deviation of the gray value difference between a pixel and all pixels in the preset neighborhood; Represents an exponential function with a natural constant as base.

[0026] In the calculation formula of the edge coefficient of the bone joint surface, The average grayscale difference between a pixel and every other pixel in the preset neighborhood The larger the The more likely a pixel is to be an edge pixel, the more likely it is that the grayscale value distribution of other pixels in the preset neighborhood is The more stable, the The greater the possibility that the pixel is an edge pixel, the greater the probability that the The greater the bone joint surface edge coefficient of the pixel point, the greater the The closer the gray value of a pixel is to 255, the more likely it is that The pixels constitute the surface of the bone joint. The larger the bone joint surface edge coefficient is, the greater the pixel point is.

[0027] In one embodiment of the present invention, the preset neighborhood is set as the area consisting of each pixel and the surrounding 8 pixels, and unless otherwise specified below, the preset neighborhoods are constructed according to the same conditions. It should be noted that in other embodiments of the present invention, the preset neighborhood can be set voluntarily, and is not limited here.

[0028] The pixel points are screened according to the bone joint surface edge coefficient of the pixel points to obtain the bone joint area. Preferably, in one embodiment of the present invention, the method for obtaining the bone joint area includes: taking the pixel points whose bone joint surface edge coefficient is greater than a preset first threshold as the bone joint surface pixel points; and taking the area composed of all the bone joint surface pixel points as the bone joint area.

[0029] In one embodiment of the present invention, the preset first threshold is set to 0.8. It should be noted that in other embodiments of the present invention, the preset first threshold can be set arbitrarily, and is not limited here.

[0030] Since most of the micro fractures are tiny gaps, the features are not obvious in the bone joint X-ray image, and the edges are fuzzy. If the bone joint edge is taken as a strong edge, the edge of the micro fracture is more likely to be a weak edge. Therefore, in the embodiment of the present invention, the strong edge in the bone joint area is weakened and the weak edge is enhanced. The fracture line of the micro fracture exists in the enhanced weak edge. Since the gradient direction of the edge pixel points of the weak edge is usually inconsistent and the pixel points are not clear, the weak edge coefficient of the edge area in the bone joint area is obtained according to the position distribution and gradient distribution of the edge pixel points in the bone joint area.

[0031] Preferably, in one embodiment of the present invention, the method for obtaining the weak edge coefficient includes: obtaining all edge areas composed of all edge pixel points in the bone joint area according to an edge detection algorithm. In one embodiment of the present invention, the canny edge detection algorithm is used to obtain the edge pixel points of the bone joint area, and the area composed of all edge pixel points is used as the edge area.

[0032] Analyze each edge area and obtain the weak edge coefficient according to the weak edge coefficient calculation formula. The weak edge coefficient calculation formula is as follows: In the formula, Indicates the weak edge coefficient of the edge region in the bone joint region; Indicates the maximum number of pixels with the same and continuous gradient direction in the edge area; Indicates the number of edge pixels in the edge area; Represents the mean gradient value of other pixels in the edge area except for the pixels with the same and continuous gradient direction; Represents the horizontal coordinate of the edge starting point in the edge area; The ordinate of the edge starting point in the edge region; The horizontal coordinate of the edge end point in the edge area; The ordinate of the edge end point in the edge region; Represents an exponential function with a natural constant as base.

[0033] In the weak edge coefficient calculation formula, the ideal state of the edge area is a straight line, that is, the value of the number of pixels in the edge area is close to So the straight-line distance The closer it is to 1, the closer the number of pixels in the edge area is to ,at this time The smaller it is, the more likely the edge area is a strong edge, so when When is larger, the edge region is more likely to be a weak edge, that is, the weak edge coefficient of the edge region is larger; when the gradient direction in the edge region is consistent and the number of continuous pixels is larger, it means that the edge region is more likely to be a strong edge, and The smaller it is, the more pixels are in the same and continuous direction. The smaller the average gradient value of other pixels in the edge area except for the pixels with the same and continuous gradient direction, the smaller the gradient value of the pixels in the edge area is. In this case, the edge area is more likely to be a weak edge, that is, the larger the weak edge coefficient of the edge area.

[0034] Step S3: Obtain the outer edge pixel points in each edge area according to the position distribution of the pixel points in the bone joint area; obtain all the outer adjacent pixel points of the outer edge pixel points in each edge area; obtain the number of fused pixel points in the preset neighborhood of each outer edge pixel point in each edge area according to the grayscale difference between each outer edge pixel point in each edge area and each outer adjacent pixel point in the preset neighborhood; obtain the fusion coefficient of each edge area according to the number of fused pixel points and the weak edge coefficient of the outer edge pixel point.

[0035] Since the edge of a tiny fracture is not easy to distinguish from the adjacent edge area, an embodiment of the present invention enhances the edge area by fusing the outer edge pixels of the edge area with the outer adjacent pixels. Since the directions of the edge pixels in the edge area of ​​a weak edge are not consistent and continuous, the degree of fusion with the outer adjacent pixels is lower than that of the edge pixels in the edge area of ​​a strong edge. Therefore, an embodiment of the present invention analyzes the fusion coefficient of the edge area by the number of pixels that can be fused in the edge area.

[0036] Preferably, in one embodiment of the present invention, the method for obtaining outer edge pixel points includes: since the outermost edge pixel points of the edge area are adjacent to the pixel points outside the edge area, in the edge area, the corresponding edge pixel points whose number of pixel points other than the edge pixel points in a preset neighborhood is not 0 are taken as outer edge pixel points.

[0037] Preferably, in one embodiment of the present invention, the method for obtaining the number of fused pixel points includes: analyzing each outer edge pixel point in each edge area, and obtaining the number of fused pixel points according to a fused pixel point number calculation formula, and the fused pixel point number calculation formula is as follows: In the formula, Indicates the edge area The number of fused pixels within the preset neighborhood of the outer edge pixels; Indicates the edge area Gray value of the outer edge pixel; Indicates the edge area The first pixel in the preset neighborhood of the outer edge pixel The gray value of each pixel; Indicates the number of edge pixels in the edge area; Indicates the edge area Gray value of edge pixels; represents the sum function; Represents an Iverson bracket, where when the condition inside the bracket is met, the value of the bracket is 1, and when the condition inside the bracket is not met, the value of the bracket is 0.

[0038] In the calculation formula of the number of fused pixels, when The outer edge pixel and the The gray value difference of pixels is not greater than When the difference between the gray value of the outer edge pixel and the mean gray value of the edge pixel is calculated, the first Pixels and The grayscale difference of the outer edge pixels is very small, so it can be considered that The neighborhood around the pixel The pixels are the pixels that can be fused. The number of pixels that can be fused in the preset neighborhood of the outer edge pixels is accumulated to obtain the The number of fused pixels within the preset neighborhood of the outer edge pixels.

[0039] Preferably, in one embodiment of the present invention, the method for obtaining the fusion coefficient includes: accumulating and normalizing the number of fused pixel points of each outer edge pixel point to obtain the fusion weight of the edge area; and taking the product of the weak edge coefficient of the edge area and the fusion weight as the fusion coefficient of the edge area. In the embodiment of the present invention, each edge area is analyzed, and the fusion coefficient calculation formula is as follows: In the formula, Represents the fusion coefficient of the edge area; The weak edge coefficient representing the edge area; Indicates the number of outer edge pixels in the edge area; Indicates the edge area The number of fused pixels within the preset neighborhood of the outer edge pixels; Indicates the number of edge pixels in the edge area.

[0040] In the fusion coefficient calculation formula, the total number of fused pixels of all outer edge pixels in the edge area is The number of edge pixels in the edge area Normalized as the weight of the weak edge coefficient of the edge area, where the larger the weak edge coefficient of the edge area, the more difficult it is to distinguish the edge area from the outer adjacent area. The larger it is, the more fused pixels there are in the edge area, and the larger the fusion coefficient of the edge area is.

[0041] Step S4: stratify the bone joint area to obtain each bone joint stratification; locate the area where the microfracture is located according to the fusion coefficient of all edge areas in each bone joint stratification.

[0042] In order to avoid the possibility that the fusion coefficients of different edge areas are too different, which may result in the inability to accurately locate the microfracture area, the embodiment of the present invention divides the bone joint area into layers.

[0043] Preferably, in one embodiment of the present invention, the method for obtaining bone joint stratification includes: stratifying the bone joint area according to a preset stratification threshold to obtain all bone joint stratifications, wherein the preset stratification threshold is set to 100, that is, the bone joint area is divided into 100 layers. It should be noted that the preset stratification threshold can be set voluntarily and is not limited here.

[0044] Preferably, in one embodiment of the present invention, the area where the microfracture is located is located according to the fusion coefficient of the edge area in each bone joint layer, including: when a certain edge area passes through a large number of bone joint layers, the edge area is more likely to be a microfracture area compared to the edge area with a smaller size. In order to avoid missing such edge areas, the product of the fusion coefficient of the edge area and the number of bone joint layers passed by the edge area is normalized as the microfracture location score of the edge area; the edge area with a microfracture location score greater than the preset second threshold is taken as the area where the microfracture is located. In one embodiment of the present invention, the preset second threshold is set to 0.9. It should be noted that the preset second threshold can be set by itself and is not limited here.

[0045] At this point, the identification of the microfracture area is completed.

[0046] In summary, the bone joint X-ray image of the patient is obtained; the bone joint surface edge coefficient of each pixel in the bone joint X-ray image is obtained according to the grayscale distribution and gradient distribution of the pixels in the bone joint X-ray image; the bone joint area is obtained according to the bone joint surface edge coefficient; all edge areas in the bone joint area are obtained; the weak edge coefficient of each edge area in the bone joint area is obtained according to the position distribution and gradient distribution of the edge pixels in the bone joint area; the outer edge pixels in each edge area are obtained according to the position distribution of the pixels in the bone joint area; all the outer adjacent pixels of the outer edge pixels in each edge area are obtained; the number of fused pixels in the preset neighborhood of each outer edge pixel in each edge area is obtained according to the grayscale difference between each outer edge pixel in each edge area and each outer adjacent pixel in the preset neighborhood; the fusion coefficient of each edge area is obtained according to the number of fused pixels and the weak edge coefficient of the outer edge pixel; the bone joint area is stratified to obtain each bone joint stratification; the area where the microfracture is located is located according to the fusion coefficient of all edge areas in each bone joint stratification.

[0047] An embodiment of the present invention provides an automatic identification system for microfractures in bone and joint X-rays based on edge detection, the system comprising a memory, a processor and a computer program, wherein the memory is used to store the corresponding computer program, the processor is used to run the corresponding computer program, and when the computer program runs in the processor, it can implement the method described in steps S1-S4.

[0048] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0049] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A method for automatically identifying microfractures in bone and joint X-rays based on edge detection, characterized in that: The method comprises: obtaining a bone joint X-ray image of a patient; obtaining a bone joint surface edge coefficient of each pixel point in the bone joint X-ray image according to the grayscale distribution and gradient distribution of the pixel points in the bone joint X-ray image; obtaining a bone joint area according to the bone joint surface edge coefficient; obtaining all edge areas in the bone joint area; obtaining a weak edge coefficient of each edge area in the bone joint area according to the position distribution and gradient distribution of the edge pixel points in the bone joint area; obtaining an outer edge pixel point in each edge area according to the position distribution of the pixel points in the bone joint area; obtaining all outer adjacent pixel points of the outer edge pixel point in each edge area; obtaining a weak edge coefficient of each outer edge pixel point in each edge area according to the position distribution of the pixel points in the bone joint ...; obtaining a weak edge coefficient of each outer edge pixel point in each edge area; obtaining a weak edge coefficient of each outer edge pixel point in each edge area; obtaining a weak edge coefficient of each outer edge pixel point in each edge area; obtaining a weak edge coefficient of each outer edge pixel point in each edge area; obtaining a weak edge coefficient of each outer edge pixel point in each edge area The grayscale difference between the pixel point and each outer adjacent pixel point in the preset neighborhood is used to obtain the number of fused pixels in the preset neighborhood of each outer edge pixel point in each edge area; the fusion coefficient of each edge area is obtained according to the number of fused pixels and the weak edge coefficient of the outer edge pixel point; the bone joint area is layered to obtain each bone joint layer; the area where the microfracture is located is located according to the fusion coefficient of all edge areas in each bone joint layer; the method for obtaining the weak edge coefficient includes: obtaining all edge areas composed of all edge pixels in the bone joint area according to the edge detection algorithm; obtaining the weak edge coefficient according to the weak edge coefficient calculation formula, and the weak edge coefficient calculation formula is as follows: In the formula, Indicates the weak edge coefficient of the edge region in the bone joint region; Indicates the maximum number of pixels with the same and continuous gradient direction in the edge area; Indicates the number of edge pixels in the edge area; Represents the mean gradient value of other pixels in the edge area except for the pixels with the same and continuous gradient direction; Represents the horizontal coordinate of the edge starting point in the edge area; The ordinate of the edge starting point in the edge region; The horizontal coordinate of the edge end point in the edge area; The ordinate of the edge end point in the edge region; Represents an exponential function with a natural constant as the base; the method for obtaining the outer edge pixel point includes: in the edge area, taking the corresponding edge pixel points whose number of pixel points other than the edge pixel points in the preset neighborhood is not 0 as the outer edge pixel points.

2. The method for automatically identifying microfractures in bone and joint X-rays based on edge detection according to claim 1, characterized in that: The method for obtaining the bone joint surface edge coefficient includes: obtaining the bone joint surface edge coefficient according to a bone joint surface edge coefficient calculation formula, and the bone joint surface edge coefficient calculation formula is as follows: In the formula, Indicates the first The edge coefficient of the bone joint surface at each pixel; Indicates the first The gradient value of each pixel; Indicates the number of other pixels in the preset neighborhood of each pixel in the bone joint X-ray image; Indicates the first The gray value of each pixel; Indicates the first The preset neighborhood of pixels The gray value of each pixel; Indicates the first The standard deviation of the gray value difference between a pixel and all pixels in the preset neighborhood; Represents an exponential function with a natural constant as base.

3. The method for automatically identifying microfractures in bone and joint X-rays based on edge detection according to claim 1, characterized in that: The method for acquiring the bone joint area includes: taking pixel points whose bone joint surface edge coefficient is greater than a preset first threshold as bone joint surface pixel points; and taking the area composed of all the bone joint surface pixel points as the bone joint area.

4. The method for automatically identifying microfractures in bone and joint X-rays based on edge detection according to claim 1, characterized in that: The method for obtaining the number of fused pixel points includes: obtaining the number of fused pixel points according to a calculation formula for the number of fused pixel points, and the calculation formula for the number of fused pixel points is as follows: In the formula, Indicates the edge area The number of fused pixels within the preset neighborhood of the outer edge pixels; Indicates the edge area Gray value of the outer edge pixel; Indicates the edge area The first pixel in the preset neighborhood of the outer edge pixel The gray value of each pixel; Indicates the number of edge pixels in the edge area; Indicates the edge area Gray value of edge pixels; represents the sum function; Represents an Iverson bracket, where when the condition inside the bracket is met, the value of the bracket is 1, and when the condition inside the bracket is not met, the value of the bracket is 0.

5. The method for automatically identifying microfractures in bone and joint X-rays based on edge detection according to claim 1, characterized in that: The method for obtaining the fusion coefficient includes: accumulating and normalizing the number of fused pixel points of each outer edge pixel point to obtain the fusion weight of the edge area; and taking the product of the weak edge coefficient of the edge area and the fusion weight as the fusion coefficient of the edge area.

6. The method for automatically identifying microfractures in bone and joint X-rays based on edge detection according to claim 1, characterized in that: The method for obtaining bone joint stratification includes: stratifying the bone joint area according to a preset stratification threshold to obtain all bone joint stratifications.

7. The method for automatically identifying microfractures in bone and joint X-rays based on edge detection according to claim 1, characterized in that: The area where the microfracture is located is located according to the fusion coefficient of the edge area in each bone joint layer, including: normalizing the product of the fusion coefficient of the edge area and the number of bone joint layer layers passed by the edge area as the microfracture location score of the edge area; and taking the edge area with a microfracture location score greater than a preset second threshold as the area where the microfracture is located.

8. An automatic identification system for microfractures in bone and joint X-rays based on edge detection, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method for automatically identifying microfractures in bone and joint X-rays based on edge detection as described in any one of claims 1 to 7 are implemented.

Images