Endoscope robot support defect detection method based on machine vision

Abstract

The invention relates to the technical field of image processing, in particular to an endoscope robot support defect detection method based on machine vision, which comprises the following steps: acquiring a foreground support image of an endoscope robot and segmenting the foreground support image into a plurality of areas; screening out a crack connected domain from the connected domains in the region according to the possibility when the connected domains in the region serve as the crack connected domain; obtaining a plurality of crack splicing areas according to direction similarity and endpoint connectivity between adjacent crack connected domains; according to the distribution of skeleton pixel points of the crack splicing area and edge pixel points on the contour of the crack splicing area, obtaining the putting number of seed points in the crack splicing area; according to the putting number of the seed points in the crack splicing area, a final crack area is obtained; and determining the quality of the endoscopic stent according to the final crack area. According to the method, the foreground bracket image of the endoscope robot is analyzed, so that the defect detection accuracy is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to an endoscopic robot stent defect detection method based on machine vision. Background Art

[0002] Endoscopic robot stents are structures or devices used to support and position endoscopic robots. The stents are designed to ensure that the robot can perform tasks accurately and stably and provide doctors with clear visualization of the patient's internal structures. However, cracks may occur on the stent surface due to material fatigue, excessive stress or other damage, which may affect the strength and durability of the stent.

[0003] Currently, when detecting crack defects in endoscopic robot stents through machine vision, region growing technology is generally used to extract the crack area. However, improper selection of the seed point position may cause the growth result to contain other surface impurities and noise, reducing the accuracy of detection. Summary of the invention

[0004] The present invention provides an endoscopic robot stent defect detection method based on machine vision to solve the existing problems.

[0005] The machine vision-based endoscopic robot stent defect detection method of the present invention adopts the following technical solutions: An embodiment of the present invention provides an endoscopic robot stent defect detection method based on machine vision, the method comprising the following steps: Acquire a foreground stent image of the endoscopic robot and segment the foreground stent image into a plurality of regions; According to the possibility that the connected domain of the region is the connected domain of the fracture, the fracture connected domain is selected from the connected domains of all regions; According to the direction similarity and endpoint connectivity between adjacent fracture connection domains, a splicing evaluation between adjacent fracture connection domains is obtained; according to the splicing evaluation between adjacent fracture connection domains, the fracture connection domains are merged to obtain a number of fracture splicing regions; The cracking degree of the crack splicing area is obtained according to the distribution of the skeleton pixel points of the crack splicing area and the edge pixel points on the outline of the crack splicing area; According to the cracking degree of the crack splicing area, the number of seed points placed in the crack splicing area is obtained; The final crack area is obtained according to the number of seed points placed in the crack splicing area; the quality of the endoscope bracket is determined according to the final crack area.

[0006] Furthermore, the foreground support image of the endoscopic robot is obtained and the foreground support image is divided into a plurality of regions, including the following specific steps: The camera captures an image of the surface of the endoscopic robot support, and performs semantic segmentation on the image to obtain a number of regions.

[0007] Furthermore, the method of selecting the fracture connected domains from the connected domains in the region according to the possibility that the connected domains in the region are used as fracture connected domains includes the following specific formula: ; in, It indicates the possibility that the connected domain of any region is a fracture connected domain. represents the maximum gray value of the connected domain, represents the average gray value of the connected domain, Indicates the first The gray value of a pixel, Indicates the number of pixels contained in the connected domain. represents the skeleton length corresponding to the connected domain, represents the maximum skeleton length of all connected domains, represents the minimum skeleton length of all connected domains, represents the length of the major axis of the minimum circumscribed ellipse of the connected domain, represents the length of the minor axis of the minimum circumscribed ellipse of this connected domain, express function; According to the possibility that the connected domain is a fracture connected domain, the fracture connected domain is screened out.

[0008] Furthermore, the method of screening out the fracture connected domains according to the possibility that the connected domains are fracture connected domains includes the following specific steps: The possibility of treating the connected domain as the fracture connected domain Greater than The connected domain of is marked and recorded as the crack connected domain. is the preset target threshold.

[0009] Furthermore, the splicing evaluation between adjacent fracture connected domains is obtained according to the directional similarity and endpoint connectivity between adjacent fracture connected domains, and the specific formula included is as follows: ; in, Represents the fracture connectivity domain Connected to the crack Evaluation of splicing, Represents the fracture connectivity domain of The largest principal component vector, Represents the fracture connectivity domain of The largest principal component vector, Represents the fracture connectivity domain The grayscale mean, Represents the fracture connectivity domain The grayscale mean, Indicates endpoint connectivity, represents the linear normalization function, Indicates directional similarity.

[0010] Furthermore, the specific calculation formula of the endpoint connectivity is as follows: ; in, Represents the fracture connectivity domain The left endpoint width, Represents the fracture connectivity domain The right endpoint width, Represents the fracture connectivity domain The left endpoint width, Represents the fracture connectivity domain The right endpoint width.

[0011] Furthermore, the step of merging the fracture connection domains to obtain a plurality of fracture connection regions according to the splicing evaluation between adjacent fracture connection domains includes the following specific steps: The splicing evaluation of the connected domain is greater than The crack connected domains are image stitched to obtain the crack stitching area. is the preset judgment threshold.

[0012] Furthermore, the cracking degree of the crack splicing area is obtained according to the distribution of the skeleton pixel points of the crack splicing area and the edge pixel points on the outline of the crack splicing area, and the specific formula included is as follows: ; in, Crack joint area The degree of cracking, The first The Euclidean distance between an edge pixel and its nearest skeleton pixel, Crack joint area The number of edge pixels, Crack joint area The skeleton length, represents the maximum skeleton length of all crack regions in the foreground scaffold image, Represents the minimum skeleton length of all crack splicing regions in the foreground scaffold image.

[0013] Furthermore, the specific formula for obtaining the number of seed points in the crack splicing area according to the cracking degree of the crack splicing area is as follows: ; in, For the The number of seed points placed in the crack splicing area, is the minimum value of the cracking degree of all crack joint areas, For the The degree of cracking in the joint area of ​​each crack is Indicates rounding down.

[0014] Furthermore, the method of obtaining a final crack area according to the number of seed points placed in the crack splicing area and determining the quality of the endoscope support according to the final crack area includes the following specific steps: According to the number of seed points in each crack splicing area, seed points are placed and regional growth is performed to obtain several candidate areas. The candidate area with the smallest grayscale mean is recorded as the final crack area. is the preset detection threshold; When the number of final crack areas is greater than , the quality of the endoscope stent is unqualified; When the number of final crack areas is less than or equal to , the quality of the endoscope stent is qualified.

[0015] The beneficial effect of the technical solution of the present invention is: this solution optimizes the problem of inaccurate growth results caused by the irregularity of the seed point placement when extracting the crack area of ​​the robot bracket through regional growing. The surface image of the robot bracket is divided into multiple regions to better capture the details in the image, and then the possible crack areas are determined by analyzing the grayscale and morphological characteristics of the cracks in the region, and then the corresponding cracking degree is calculated after splicing to finally determine the seed point placement position and number, so that the seed points are placed in line with the crack location, thereby improving the accuracy of defect detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be 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.

[0017] Figure 1 The present invention is a flowchart of the steps of the endoscopic robot stent defect detection method based on machine vision. DETAILED DESCRIPTION

[0018] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation, structure, features and effects of the endoscope robot stent defect detection method based on machine vision 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.

[0019] 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.

[0020] The specific scheme of the endoscope robot stent defect detection method based on machine vision provided by the present invention is described in detail below with reference to the accompanying drawings.

[0021] See also Figure 1 , which shows a flowchart of a method for detecting defects in an endoscopic robot stent based on machine vision according to an embodiment of the present invention, the method comprising the following steps: Step S001: Acquire a foreground support image of the endoscopic robot and divide the foreground support image into several regions.

[0022] The image of the surface of the endoscopic robot stent is captured by a high-resolution camera to ensure that all surface areas of the robot stent can be captured by the camera, and then the captured image is grayed and subjected to Gaussian filtering and noise reduction processing to obtain a grayscale image. In order to improve the accuracy of crack defect recognition, this embodiment uses a semantic segmentation neural network to segment the grayscale image to obtain the semantic area of ​​the endoscopic robot stent. All pixels within the semantic area on the grayscale image constitute the foreground stent image. Irrelevant areas such as the background in the grayscale image are not considered. The semantic segmentation network used in this embodiment is DeepLabV3.

[0023] Step S002: screening out fracture connected domains from the connected domains in the region according to the possibility that the connected domains in the region are fracture connected domains.

[0024] It should be noted that for the foreground support image, there may be some relatively subtle cracks in the support, which may result in the inability to accurately extract the subtle cracks based on the analysis of the entire foreground support image. Therefore, this embodiment divides the foreground support image into multiple regions using the SLIC algorithm, and each region is a connected domain. Each region is then analyzed, and in each region, the connected domain that may belong to the crack is screened out based on the morphology and grayscale characteristics of the connected domain. The disconnected cracks caused by the regional segmentation are then spliced ​​to obtain a complete crack region.

[0025] Specifically, the SLIC algorithm is used to divide the foreground area image of the endoscope robot stent into multiple areas. The SLIC algorithm is a well-known technology and will not be described in detail in this embodiment. Since the light receiving area in the crack area on the surface of the endoscope robot stent is small, its grayscale is deeper and smoothly distributed than the grayscale on the stent surface. At the same time, most of the formation of cracks is caused by excessive stress. Therefore, the shape of the cracks is generally slender and extends in a certain direction, while other stains or texture areas on the stent surface may not have such a shape. Therefore, based on the above analysis, the calculation formula for the possibility that any area is a crack connected domain is obtained as follows: ; in, It indicates the possibility that the connected domain of any region is a fracture connected domain. represents the maximum gray value of this connected domain, represents the average gray value of this connected domain, Indicates the first The gray value of a pixel, Indicates the number of pixels contained in the connected domain. Indicates the length of the skeleton corresponding to this connected domain, where the skeleton corresponding to the connected domain is obtained using the K3M algorithm. represents the maximum skeleton length of all connected domains, represents the minimum skeleton length of all connected domains, represents the length of the major axis of the minimum circumscribed ellipse of this connected domain, represents the length of the minor axis of the minimum circumscribed ellipse of this connected domain, express function.

[0026] is the average difference between the maximum grayscale value of pixels in the connected domain and the grayscale values ​​of all pixels in the connected domain, which is used to indicate the uniformity of grayscale distribution in this connected domain. The smaller the value of is, the more concentrated the gray level of the pixels in the connected domain is, that is, the gray level is more uniform and smooth, and the connected domain is more likely to be a crack connected domain. Represents the gray level of the entire connected domain. Since the gray level of the crack area is darker than that of the stent surface, The smaller the value is, the darker the overall grayscale of this connected domain is, and the more likely it is a crack connected domain. represents the relative skeleton length of this connected domain. Since the crack connected domain presents an extended form, the corresponding morphological skeleton is obtained for the connected domain. The length of the skeleton pixel represents the extension degree of this connected domain. Therefore, if The larger the value of is, the longer the extension length of this connected domain is, and the more likely it is to be a crack connected domain. represents the slender characteristics of the connected domain. Since the fracture connected domain is overall slender, The degree of this slender shape can be reflected. The larger the value of is, the more likely this connected domain is to have a slender morphological feature, and the more likely it is to be a fracture connected domain. In summary, the screening index of fracture connected domain is obtained. , calculate any connected domain in all regions and use Function normalized to within the range.

[0027] According to the above method, the probability of all connected domains being fracture connected domains is obtained, and the target threshold is preset. ,Will Greater than The connected domain of is marked and recorded as the crack connected domain.

[0028] At this point, the connected domain of the crack is obtained.

[0029] Step S003: obtaining a splicing evaluation between adjacent fracture connected domains according to the directional similarity and endpoint connectivity between adjacent fracture connected domains, and merging the fracture connected domains according to the splicing evaluation between adjacent fracture connected domains to obtain a plurality of fracture splicing regions.

[0030] It should be noted that when screening possible crack connection domains in the segmented area, some longer cracks are divided into multiple areas to form disconnected cracks. In order to improve the integrity of crack extraction, this embodiment stitches multiple areas into a complete foreground image. However, since there may be errors in stitching multiple areas, some disconnected cracks are misplaced and cannot be accurately stitched. Therefore, this embodiment analyzes the crack connection domains in the area, establishes stitching indicators to accurately stitch the disconnected cracks to obtain all complete crack areas, and then calculates the corresponding cracking degree based on the morphology of the complete crack area.

[0031] Specifically, the disconnected cracks are first spliced. The multiple cracks that need to be spliced ​​are originally a natural crack. Due to image segmentation, they are disconnected but their relative positions remain unchanged. Therefore, the multiple crack values ​​that need to be spliced ​​will only exist in any adjacent areas. For cracks, the multiple cracks that need to be spliced ​​will have similar extension directions, and their internal grayscale distributions are similar. In addition, because the two cracks to be spliced ​​are spliced ​​head to tail, that is, for any disconnected crack in a certain area, its left end is spliced ​​with the right end of the corresponding crack in the left adjacent area, and its right end is spliced ​​with the left end of the corresponding crack in the right adjacent area. The more similar the widths of the corresponding endpoints of the two cracks are, the higher the possibility of splicing the two cracks. Therefore, based on the above characteristics, the crack connectivity domain in a certain area is determined. Domains connected to cracks in any adjacent region The calculation formula for the splicing evaluation is: ; in, Represents the fracture connectivity domain Connected to the crack Evaluation of splicing, Represents the fracture connectivity domain of The largest principal component vector, Represents the fracture connectivity domain of The largest principal component vector, Represents the fracture connectivity domain The grayscale mean of Represents the fracture connectivity domain The grayscale mean, Represents the fracture connectivity domain The left endpoint width, Represents the fracture connectivity domain The right endpoint width, Represents the fracture connectivity domain The left endpoint width, Represents the fracture connectivity domain The right endpoint width, represents the linear normalization function.

[0032] It should be noted that, in this embodiment, in the fracture connection domain In the direction of the maximum principal component vector, the first pixel point on the crack connected domain passed clockwise is the left endpoint, and the last pixel point passed clockwise is the right endpoint.

[0033] The calculation method of the width of the endpoint in this embodiment is: on the contour edge of the crack connected domain, obtain the Nq edge pixels closest to the left endpoint (including the left endpoint), and the left side length of the minimum circumscribed rectangle of these Nq edge pixels is used as the width of the left endpoint. On the contour edge of the crack connected domain, obtain the Nq edge pixels closest to the right endpoint (including the right endpoint), and the right side length of the minimum circumscribed rectangle of these Nq edge pixels is used as the width of the right endpoint. This embodiment is described by taking Nq=20 as an example, and other values ​​can be set in other embodiments, which are not limited in this embodiment.

[0034] Represents the difference in angle between the corresponding vectors of the two cracks, indicating the similarity of direction, through The maximum principal direction vector represents the extension direction of the crack. Map the angle to a positive value to reduce the calculation error caused by negative values. The larger the value is, the smaller the angle between the two cracks is, and the more consistent the extension directions of the two cracks are. Therefore, the greater the possibility of being a disconnected crack, the greater the possibility of splicing the two cracks. It represents the average grayscale difference between the two crack areas. If the two cracks are disconnected cracks, the corresponding grayscale difference will be smaller. The smaller the value, the greater the possibility of the two cracks joining. Represents the difference in width between the two end points of the two crack regions, indicating the connectivity of the endpoints. Since the end-to-end splicing is adopted, for the crack connected domain For example, this embodiment is based on The extension direction is taken as the horizontal direction, then if the crack connects the domain In the fracture connection area In the left area of ​​, then compare The left endpoint width and The left endpoint width If the widths of the two are similar, The smaller the size, the more likely it is that the two originally belonged to one crack, so the possibility of splicing is greater. In the fracture connection area In the right area of The smaller the value, the greater the possibility of splicing the two crack areas. Therefore, if the splicing conditions are met, and There will be any extremely small value in , then The smaller the value is, the greater the possibility of splicing the corresponding two crack regions. Finally, for any crack connected domain The fracture connection domain in the adjacent area Calculate the corresponding The values ​​are linearly normalized to the range [0,1], and 1 is added to the denominator to avoid the denominator being zero.

[0035] It should be noted that all mathematical models in this embodiment are preferred implementation methods. In other embodiments, when the denominator is 0, the denominator can be increased by 1 based on the mathematical model provided in this embodiment to avoid the problem of non-implementation due to the denominator being 0.

[0036] According to the above method, the splicing evaluation of any two connected domains is obtained, and the judgment threshold is preset. , the splicing evaluation of the connected domain Greater than All connected domains of the crack are marked, and the marked crack connected domains are image stitched to obtain a complete crack area, which is recorded as the crack stitching area.

[0037] At this point, the crack joint area is obtained.

[0038] Step S004: Obtain the cracking degree of the crack splicing area according to the distribution of the skeleton pixel points of the crack splicing area and the edge pixel points on the outline of the crack splicing area.

[0039] It should be noted that the placement position of the regional growth seed points is determined according to the position of the obtained crack splicing area, but since the sizes and lengths of the cracks are different, an equal number of seed points are placed in all crack splicing areas and regional growth is performed. If the number of seed points is small, large cracks may be incomplete, and if the number of seed points is large, small cracks may cause excessive growth. To address this problem, this embodiment calculates the degree of cracking in each crack splicing area, and then determines the number of different seed points to be placed according to the degree of cracking, so that the growth result is more accurate.

[0040] It is further explained that for the crack splicing area, the wider the width of the crack and the longer the extension length of the crack, the greater the degree of cracking. To measure the extension length of a crack, the skeleton length in the morphological skeleton can be used. The skeleton length represents the overall trunk feature of this area. Of course, the skeleton pixels can also show width characteristics. Since the skeleton pixels are inside or at the boundary of the connected domain and the crack widths at different positions are different, the crack width is measured by calculating the Euclidean distance between the edge point pixels of the crack splicing area and the nearest skeleton pixels.

[0041] Specifically, the calculation formula for the cracking degree of the crack joint area is: ; in, Crack joint area The degree of cracking, The first The Euclidean distance between an edge pixel and its nearest skeleton pixel, Crack joint area The number of edge pixels, Crack joint area The skeleton length, represents the maximum skeleton length of all crack regions in the entire image, Indicates the maximum skeleton length and minimum skeleton length of all crack splicing areas in the entire image.

[0042] Represents the crack width, which is measured by the average statistical value of the closest distance between the edge pixel point and the skeleton pixel point in the crack splicing area. If the average statistical value is larger, it means that the overall crack width is larger, then the crack splicing area The greater the degree of cracking; Represents the crack joint area The relative skeleton length of the crack joint area The longer the skeleton length is among all the skeleton lengths of the crack area, that is, the longer the extension length of the crack is, the greater the cracking degree is. Combining the crack width and the extension length, the cracking degree of any crack splicing area can be obtained. .

[0043] According to the above method, the cracking degree of all crack joint areas is obtained.

[0044] At this point, the cracking extent of all crack joint areas is obtained.

[0045] Step S005: According to the cracking degree of the crack splicing area, the number of seed points placed in the crack splicing area is obtained.

[0046] It should be noted that in order to make the edge of the crack splicing area more accurate, this embodiment uses regional growing technology to determine the seed point placement rules based on the position of the crack splicing area and the corresponding cracking degree to perform regional growth, thereby growing an accurate and complete crack splicing area image.

[0047] Specifically, for the placement of seed points, the seed points can be evenly placed according to the location of each crack splicing area, and the number of seed points placed is determined by the degree of cracking in the crack splicing area. If the degree of cracking in a crack splicing area is greater, the more serious the crack is, and the more seed points are placed. The calculation formula for the number of seed points placed is as follows: ; in, For the The number of seed points placed in the crack splicing area, is the minimum value of the cracking degree of all crack joint areas, For the The degree of cracking in the joint area of ​​each crack is Indicates rounding down.

[0048] According to the above method, the number of seed points placed in each crack joint area is obtained.

[0049] At this point, the number of seed points placed in each crack splicing area is obtained.

[0050] Step S006: Obtain a final crack area according to the number of seed points placed in the crack splicing area; and determine the quality of the endoscope support according to the final crack area.

[0051] According to the number of seed points in each crack splicing area, seed points are placed and regional growth is performed to obtain several candidate areas, that is, each crack splicing area is divided into several candidate areas, and the candidate area with the smallest grayscale mean in each crack splicing area is recorded as the final crack area, wherein the regional growing algorithm is a prior art and will not be repeated here. After obtaining the final crack area, the crack position and the corresponding cracking degree are marked on the corresponding endoscope robot bracket surface image. The preset detection threshold of this embodiment , other embodiments may be set to other values, which are not limited in this embodiment; When the number of final crack areas is greater than , the quality of the endoscope stent is unqualified; When the number of final crack areas is less than or equal to , the quality of the endoscope stent is qualified.

[0052] At this point, this embodiment is completed.

[0053] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An endoscopic robot stent defect detection method based on machine vision, characterized in that: The method comprises the following steps: Acquire a foreground stent image of the endoscopic robot and segment the foreground stent image into a plurality of regions; According to the possibility that the connected domain of the region is the connected domain of the fracture, the fracture connected domain is selected from the connected domains of all regions; According to the direction similarity and endpoint connectivity between adjacent fracture connection domains, a splicing evaluation between adjacent fracture connection domains is obtained; according to the splicing evaluation between adjacent fracture connection domains, the fracture connection domains are merged to obtain a number of fracture splicing regions; The cracking degree of the crack splicing area is obtained according to the distribution of the skeleton pixel points of the crack splicing area and the edge pixel points on the outline of the crack splicing area; According to the cracking degree of the crack splicing area, the number of seed points placed in the crack splicing area is obtained; The final crack area is obtained according to the number of seed points placed in the crack splicing area; the quality of the endoscope bracket is determined according to the final crack area.

2. The method for detecting defects in an endoscopic robot stent based on machine vision according to claim 1, characterized in that: The specific steps of obtaining the foreground stent image of the endoscopic robot and dividing the foreground stent image into a plurality of regions are as follows: The camera captures an image of the surface of the endoscopic robot support, and performs semantic segmentation on the image to obtain a number of regions.

3. The method for detecting defects of an endoscopic robot stent based on machine vision according to claim 1, characterized in that: The possibility of the connected domain of the region being the connected domain of the fracture is selected from the connected domains of all regions, and the specific formula included is as follows: ； in, It indicates the possibility that the connected domain of any region is a fracture connected domain. represents the maximum gray value of the connected domain, represents the average gray value of the connected domain, Indicates the first The gray value of a pixel, Indicates the number of pixels contained in the connected domain. represents the skeleton length corresponding to the connected domain, represents the maximum skeleton length of all connected domains, represents the minimum skeleton length of all connected domains, represents the length of the major axis of the minimum circumscribed ellipse of the connected domain, represents the length of the minor axis of the minimum circumscribed ellipse of this connected domain, express function; According to the possibility that the connected domain is a fracture connected domain, the fracture connected domain is screened out.

4. The method for detecting defects in an endoscopic robot stent based on machine vision according to claim 3, characterized in that: The method of screening out the fracture connected domains according to the possibility that the connected domains are fracture connected domains comprises the following specific steps: The possibility of treating the connected domain as the fracture connected domain Greater than The connected domain of is marked and recorded as the crack connected domain. is the preset target threshold.

5. The method for detecting defects of an endoscopic robot stent based on machine vision according to claim 1, characterized in that: The splicing evaluation between adjacent fracture connected domains is obtained according to the direction similarity and endpoint connectivity between adjacent fracture connected domains, and the specific formula included is as follows: ； in, Represents the fracture connectivity domain Connected to the crack Evaluation of splicing, Represents the fracture connectivity domain of The largest principal component vector, Represents the fracture connectivity domain of The largest principal component vector, Represents the fracture connectivity domain The grayscale mean, Represents the fracture connectivity domain The grayscale mean, Indicates endpoint connectivity, represents the linear normalization function, Indicates directional similarity.

6. The method for detecting defects in an endoscopic robot stent based on machine vision according to claim 5, characterized in that: The specific calculation formula of the endpoint connectivity is as follows: ； in, Represents the fracture connectivity domain The left endpoint width, Represents the fracture connectivity domain The right endpoint width, Represents the fracture connectivity domain The left endpoint width, Represents the fracture connectivity domain The right endpoint width.

7. The method for detecting defects of an endoscopic robot stent based on machine vision according to claim 1, characterized in that: The method of merging the fracture connection domains according to the splicing evaluation between adjacent fracture connection domains to obtain a plurality of fracture splicing regions includes the following specific steps: The splicing evaluation of the connected domain is greater than The crack connected domains are image stitched to obtain the crack stitching area. is the preset judgment threshold.

8. The method for detecting defects in an endoscopic robot stent based on machine vision according to claim 1, characterized in that: The cracking degree of the crack splicing area is obtained according to the distribution of the skeleton pixel points of the crack splicing area and the edge pixel points on the outline of the crack splicing area, and the specific formula included is as follows: ； in, Crack joint area The degree of cracking, The first The Euclidean distance between an edge pixel and its nearest skeleton pixel, Crack joint area The number of edge pixels, Crack joint area The skeleton length, represents the maximum skeleton length of all crack regions in the foreground scaffold image, Represents the minimum skeleton length of all crack splicing regions in the foreground scaffold image.

9. The method for detecting defects of an endoscopic robot stent based on machine vision according to claim 1, characterized in that: The specific formula for obtaining the number of seed points in the crack splicing area according to the cracking degree of the crack splicing area is as follows: ； in, For the The number of seed points placed in the crack splicing area, is the minimum value of the cracking degree of all crack joint areas, For the The degree of cracking in the joint area of ​​each crack is Indicates rounding down.

10. The method for detecting defects of an endoscopic robot stent based on machine vision according to claim 1, characterized in that: The method of obtaining a final crack area according to the number of seed points placed in the crack splicing area and determining the quality of the endoscope stent according to the final crack area includes the following specific steps: According to the number of seed points in each crack splicing area, seed points are placed and regional growth is performed to obtain several candidate areas. The candidate area with the smallest grayscale mean is recorded as the final crack area. is the preset detection threshold; When the number of final crack areas is greater than , the quality of the endoscope stent is unqualified; When the number of final crack areas is less than or equal to , the quality of the endoscope stent is qualified.

Images