A method, device and system for detecting the processing quality of mining parts

Abstract

This application relates to the field of image detection technology, and specifically relates to a method, device and system for detecting the processing quality of mining parts. The method includes: obtaining the surface grayscale images of each orientation of the mining parts; determining the direction discreteness of each edge contour in the surface grayscale image of each orientation; determining the grayscale discreteness of each edge contour in the surface grayscale image of each orientation; determining the color feature value of each edge contour in the surface grayscale image of each orientation; determining the drift trend degree of each edge contour in the surface grayscale image of each orientation; determining the shape feature value of each edge contour in the surface grayscale image of each orientation; determining the discrimination coefficient of each edge contour in the surface grayscale image of each orientation, and detecting the processing quality of the mining parts. This application improves the accuracy of detecting surface defects of mining parts, which is beneficial to the normal operation of mining equipment and the safe production of mining operations.

Description

Technical Field

[0001] This application relates to the technical field of image detection, and particularly to a method, device and system for detecting the processing quality of mining parts. Background Art

[0002] Mining parts play a crucial role in mining operations. The processing quality of these parts not only directly affects the operation reliability and production efficiency of equipment, but also relates to the safety and stability of mining operations. Therefore, strict detection of the processing quality of mining parts is of great significance for ensuring the normal operation of equipment and safe production in mines.

[0003] When detecting the processing quality of mining parts, traditional detection methods perform threshold segmentation based on the gray value of the image to achieve defect detection. When the gray value of the defect area is close to that of the normal area, for example, when the gray values of the pixel points at the crack area and the corner of the surface of the mining part are close, missed detections will occur, resulting in a low accuracy rate of defect detection for mining parts and affecting the precision of the quality detection of mining parts. Summary of the Invention

[0004] To solve the above technical problems, a method, device and system for detecting the processing quality of mining parts are provided to solve the existing problems.

[0005] The solution of this application to solve the technical problem is to provide a method, device and system for detecting the processing quality of mining parts, including the following steps:

[0006] In the first aspect, an embodiment of this application provides a method for detecting the processing quality of mining parts, and the method includes the following steps:

[0007] Obtain the surface gray-scale images of each orientation of the mining part, and extract each edge contour in the surface gray-scale image of each orientation;

[0008] According to the discrete situation of the gradient directions of the edge pixel points on each edge contour, determine the direction discreteness of each edge contour in the surface gray-scale image of each orientation; according to the discrete situation of the distribution of the gray values of the pixel points in the area surrounded by each edge contour, determine the gray discreteness of each edge contour in the surface gray-scale image of each orientation; according to the fluctuation range of the gray values of the pixel points in the area surrounded by each edge contour, combine the direction discreteness and the gray discreteness to determine the color feature value of each edge contour in the surface gray-scale image of each orientation.

[0009] According to the curvature characteristics of the edge pixel points on each edge contour and the change of the drift direction after mean shift of the edge pixel points corresponding to the extreme values of all curvatures, determine the drift trend degree of each edge contour in the surface gray image in each direction; according to the proportion degree of the number of edge pixel points corresponding to the extreme values of all curvatures on each edge contour, and in combination with the drift trend degree, determine the shape characteristic value of each edge contour in the surface gray image in each direction;

[0010] Based on the color characteristic value and the shape characteristic value, determine the discrimination coefficient of each edge contour in the surface gray image in each direction, and based on the discrimination coefficients in all directions, detect the processing quality of the mining parts.

[0011] Preferably, the direction dispersion degree of each edge contour in the surface gray image in each direction is the variance of the gradient direction angles of all edge pixel points on each edge contour.

[0012] Preferably, the determination of the gray dispersion degree of each edge contour in the surface gray image in each direction includes:

[0013] Denote the pixel points in the area surrounded by each edge contour as internal pixel points;

[0014] Calculate the mean value of the gray values of any internal pixel point on each edge contour and all pixel points in its neighborhood, and denote it as the local gray value of the any internal pixel point;

[0015] Take the dispersion degree of the local gray values of all internal pixel points on each edge contour as the gray dispersion degree of each edge contour in the surface gray image in each direction.

[0016] Preferably, the determination of the color characteristic value of each edge contour in the surface gray image in each direction includes:

[0017] Calculate the range of the local gray values of all internal pixel points on each edge contour;

[0018] Take the product of the range, the gray dispersion degree and the direction dispersion degree as the color characteristic value of each edge contour in the surface gray image in each direction.

[0019] Preferably, the determination of the drift trend degree of each edge contour in the surface gray image in each direction includes:

[0020] Taking any edge pixel point on each edge contour as the starting point, arrange all edge pixel points on each edge contour in the clockwise direction to obtain the extreme values of the curvatures of all edge pixel points on each edge contour;

[0021] Denote all the edge pixels corresponding to the extreme values as extreme value pixels; use the position coordinates of all the extreme value pixels on each edge contour as the input of the mean shift algorithm to obtain the drift direction angle of each extreme value pixel in each iteration.

[0022] Take the mean of the drift direction angles of all the extreme value pixels on each edge contour in all iterations in the surface gray-scale image of each azimuth as the drift trend degree of each edge contour in the surface gray-scale image of each azimuth.

[0023] Preferably, determining the shape feature value of each edge contour in the surface gray-scale image of each azimuth includes:

[0024] Count the number of all the extreme value pixels on each edge contour, denoted as the first quantity;

[0025] Count the number of all the edge pixels on each edge contour, denoted as the second quantity;

[0026] Calculate the ratio of the first quantity to the second quantity, and take the product of the reciprocal of the drift trend degree and the ratio as the shape feature value of each edge contour in the surface gray-scale image of each azimuth.

[0027] Preferably, the discrimination coefficient of each edge contour in the surface gray-scale image of each azimuth is the normalized result of the product of the color feature value and the shape feature value.

[0028] Preferably, detecting the processing quality of the mining parts includes: if there is an edge contour in the surface gray-scale images of all azimuths whose discrimination coefficient is greater than the preset threshold, the mining parts are defective; otherwise, the mining parts are not defective.

[0029] In a second aspect, an embodiment of the present application further provides a mining part processing quality detection device, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the steps of the above-mentioned mining part processing quality detection method are implemented.

[0030] In a third aspect, an embodiment of the present application further provides a mining part processing quality detection system. A computer program is stored in the system, and when the computer program is executed by a processor, the steps of the above-mentioned mining part processing quality detection method are implemented.

[0031] The present application has at least the following beneficial effects:

[0032] The present application determines the direction discreteness of each edge contour in the surface gray-scale image in each orientation according to the discreteness of the gradient directions of the edge pixel points on each edge contour. The beneficial effect is that it takes into account the change of the gradient directions of the edge pixel points to reflect the possibility that the corresponding edge contour is a crack; determines the gray-scale discreteness of each edge contour in the surface gray-scale image in each orientation according to the discreteness of the distribution of the gray-scale values of the pixel points within the area enclosed by each edge contour. The beneficial effect is that it takes into account the gray-scale change of the pixel points inside the edge contour caused by the depth of the crack to reflect the distribution of the gray-scale values of the pixel points, and further indicates the possibility that the contour edge is a crack; determines the color feature value of each edge contour in the surface gray-scale image in each orientation according to the fluctuation range of the gray-scale values of the pixel points within the area enclosed by each edge contour, in combination with the direction discreteness and the gray-scale discreteness. The beneficial effect is that it combines the change of the gradient direction of the edge pixel points with the gray-scale change of the pixel points inside the edge contour to further illustrate the characteristic difference between the crack and the edge contour, so as to reflect the possibility that the edge contour is a crack; determines the drift trend degree of each edge contour in the surface gray-scale image in each orientation according to the curvature feature of the edge pixel points on each edge contour and the change of the drift direction after mean shift of the edge pixel points corresponding to the extreme values of all curvatures. The beneficial effect is that it takes into account the curvature change of the edge pixel points and analyzes the change of the drift direction of the edge pixel points corresponding to the extreme values of the curvature during each drift process to reflect the possibility that the corresponding edge pixel points are located on the crack; determines the shape feature value of each edge contour in the surface gray-scale image in each orientation according to the proportion degree of the number of the edge pixel points corresponding to the extreme values of all curvatures on each edge contour, in combination with the drift trend degree. The beneficial effect is that it takes into account the difference in the curvature change of the edge pixel points on the crack and the edge, as well as the degree of consistency of the drift direction, to further reflect the possibility that the corresponding edge pixel points are located on the crack; based on the color feature value and the shape feature value, determines the discrimination coefficient of each edge contour in the surface gray-scale image in each orientation to detect the processing quality of the mining parts. The beneficial effect is that it can more accurately detect the crack defects of the mining parts, improve the accuracy of the surface defect detection of the mining parts, and is beneficial to the normal operation of the mining equipment and the safe production of the mining operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The following further elaborates in detail a method for detecting the processing quality of mining parts according to the present application with reference to the accompanying drawings.

[0034] Figure 1 It is a step flow chart of a method for detecting the processing quality of mining parts provided by an embodiment of the present application;

[0035] Figure 2Flowchart of the steps for obtaining the color feature values of each edge contour in the surface grayscale image in each orientation provided by the embodiments of the present application;

[0036] Figure 3 Flowchart of the steps for obtaining the shape feature values of each edge contour in the surface grayscale image in each orientation provided by the embodiments of the present application. Detailed implementation manners

[0037] In order to make the objectives, technical solutions and advantages of the present application clearer, the following further elaborates in detail a method, device and system for detecting the processing quality of mining parts proposed by the present application in combination with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0038] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present application belongs.

[0039] Please refer to Figure 1 , which shows the flowchart of the steps of a method for detecting the processing quality of mining parts provided by an embodiment of the present application. The method includes the following steps:

[0040] Step 1, obtain the surface grayscale images of the mining parts in each orientation.

[0041] Place a set of CCD cameras around a mining part, collect the surface images of the mining part, and perform denoising, image enhancement, image sharpening and grayscale processing on the collected images to obtain the surface grayscale images in each orientation.

[0042] Preferably, in this embodiment, the median filtering algorithm is used to denoise the image, the histogram equalization algorithm is used for image enhancement, and the Laplace algorithm is used to sharpen the image. Among them, the median filtering algorithm, the histogram equalization algorithm and the Laplace algorithm are well-known technologies and will not be elaborated here.

[0043] So far, the surface grayscale images in each orientation are obtained.

[0044] Step 2, determine the direction dispersion of each edge contour in the surface grayscale image in each orientation according to the discrete situation of the gradient directions of the edge pixels on each edge contour; determine the gray dispersion of each edge contour in the surface grayscale image in each orientation according to the discrete situation of the distribution of the gray values of the pixels in the area surrounded by each edge contour; determine the color feature value of each edge contour in the surface grayscale image in each orientation according to the fluctuation range of the gray values of the pixels in the area surrounded by each edge contour, in combination with the direction dispersion and the gray dispersion.

[0045] During the casting process of mining parts, due to uneven heating of the casting, it is easy to cause defects in the shape of cracks in the mining parts during the production process, making the mining parts prone to breakage during use. During the production process, it is necessary to detect the mining parts with unqualified processing quality.

[0046] Furthermore, the flowchart of the steps of the method for obtaining the color feature value of each edge contour in the surface grayscale image of each orientation provided by the embodiment of the present application is as follows: Figure 2 shown.

[0047] First, by extracting the edge contours in the surface grayscale image, all the contours that may be cracks in the surface grayscale image are initially extracted, specifically:

[0048] Perform edge detection on the surface grayscale image at each orientation, extract each edge contour, and the gradient direction angle of each edge pixel point;

[0049] Preferably, in this embodiment, a canny edge detection algorithm is used for edge detection, wherein the canny edge detection algorithm is a well-known technology and will not be described in detail here.

[0050] Furthermore, all the extracted edge contours include angular contours in the normal structure of the parts and possible crack contours. In order to detect whether there is a crack contour from the edge contour, the characteristics of all edge contours are analyzed to distinguish between angular contours and crack contours.

[0051] Since the crack may be irregular, the color change of the pixels on both sides of the crack outline may be from light to dark or from dark to light, and this change is irregular; while on both sides of the angular outline, the color change is either from light to dark or from dark to light, and the color change on both sides is consistent, so the change in the gradient direction of the edge pixels on the crack outline will be more drastic than the change in the gradient direction of the edge pixels on the angular outline; secondly, in the area surrounded by the crack outline, the deeper area has a darker color and the lighter area has a lighter color, and the color distribution of the pixels within the crack outline is uneven, so the grayscale value of the pixels inside the crack outline will show a certain fluctuation.

[0052] Based on the above analysis, the gradient changes of edge pixels on each edge contour and the grayscale changes of pixels in the area surrounded by each edge contour are analyzed to determine the color feature value, which preliminarily reflects the possibility that the edge contour is a crack contour, specifically:

[0053] The variance of the gradient direction angles of all edge pixels on each edge contour in the surface grayscale image of each orientation is used as the directional discreteness of each edge contour in the surface grayscale image of each orientation;

[0054] The pixel points within the region enclosed by each edge contour are denoted as internal pixel points;

[0055] Calculate the mean value of the grayscale values of any internal pixel point of each edge contour and all pixel points within its neighborhood, and denote it as the local grayscale of the any internal pixel point;

[0056] Preferably, in this embodiment, calculate the mean value of the grayscale values of any internal pixel point of each edge contour and all pixel points within its 8-neighborhood, and denote it as the local grayscale of the any internal pixel point.

[0057] Take the degree of dispersion of the local grayscales of all internal pixel points of each edge contour as the grayscale dispersion degree of each edge contour;

[0058] Preferably, in this embodiment, take the variance of the local grayscales of all internal pixel points of each edge contour as the grayscale dispersion degree of each edge contour.

[0059] Calculate the range of the local grayscales of all internal pixel points of each edge contour, and take the product of the range, the grayscale dispersion degree, and the direction dispersion degree as the color feature value of each edge contour in the surface grayscale image of each azimuth;

[0060] Preferably, in this embodiment, the calculation method of the color feature value of each edge contour in the surface grayscale image of each azimuth is: C m,k = α m,k × V m,k × F m,k , where C m,k is the color feature value of the k-th edge contour in the surface grayscale image of the m-th azimuth, α m,k is the range of the local grayscales of all internal pixel points of the k-th edge contour in the surface grayscale image of the m-th azimuth, V m,k is the variance of the local grayscales of all internal pixel points of the k-th edge contour in the surface grayscale image of the m-th azimuth, that is, the grayscale dispersion degree, and F m,k is the variance of the gradient direction angles of all edge pixel points on the k-th edge contour in the surface grayscale image of the m-th azimuth, that is, the direction dispersion degree.

[0061] It should be noted that the larger the direction dispersion degree, the more drastic the change in the gradient direction of the edge pixel points on the corresponding edge contour, and the more likely the corresponding edge contour is a crack contour; secondly, the larger the range and the grayscale dispersion degree, the greater the difference in the grayscale values of the internal pixel points of the corresponding edge contour and the greater the fluctuation range, and the more likely the corresponding edge contour is a crack contour. Therefore, the larger the obtained color feature value, the more likely the edge contour is a crack.

[0062] So far, the color feature values of each edge contour in the surface gray-scale image in each direction are obtained.

[0063] Step 3: Determine the drift trend degree of each edge contour in the surface gray-scale image in each direction according to the curvature characteristics of the edge pixel points on each edge contour and the change of the drift direction after mean shift of the edge pixel points corresponding to the extreme values of all curvatures; Determine the shape feature value of each edge contour in the surface gray-scale image in each direction according to the proportion degree of the number of edge pixel points corresponding to the extreme values of all curvatures on each edge contour, combined with the drift trend degree.

[0064] Furthermore, the shape of the crack contour is more tortuous, while the shape of the edge contour is mostly smoother. Among the curvatures of all edge pixel points on each edge contour, the larger the proportion of the number of extreme values, the more likely it is a crack; Secondly, the distribution densities of the edge pixel points corresponding to different extreme values are different. When performing mean shift on the extreme values, the edge pixel points corresponding to different extreme values will move towards the most complex area of the crack, and the drift direction of the edge pixel points on the crack contour will not change much during each drift process; For all edge pixel points on the edge contour, even if there is wear, or due to the complexity of the mechanical structure, there are extreme values in the curvatures of multiple edge pixel points, the complexity at the edge pixel points corresponding to each extreme value is usually similar, that is, the local densities of the edge pixel points corresponding to each extreme value are close. When performing mean shift, the moving directions of the edge pixel points corresponding to each extreme value are not fixed, so the drift direction of the edge pixel points on the edge contour will change greatly during each drift process.

[0065] Furthermore, the step flow chart of the method for obtaining the shape feature value of each edge contour in the surface gray-scale image in each direction provided by the embodiment of the present application is as Figure 3 shown.

[0066] Based on the above analysis, analyze the curvature change of the edge pixel points on each edge contour, and determine the shape feature value to reflect the possibility that the corresponding edge contour is a crack contour. Specifically:

[0067] Taking any edge pixel point on each edge contour in the surface gray-scale image in each direction as the starting point, arrange all the edge pixel points on each edge contour in the clockwise direction, and obtain the extreme values of the curvatures of all the edge pixel points on each edge contour;

[0068] Preferably, in this embodiment, the difference method is used to obtain the extreme values. Among them, the difference method is a well-known technology and will not be elaborated here.

[0069] All the edge pixels corresponding to the extreme values are denoted as extreme value pixels. The position coordinates of all the extreme value pixels on each edge contour are used as the input of the mean shift algorithm to obtain the drift direction angle of each extreme value pixel in each iteration.

[0070] Preferably, in this embodiment, the mean shift algorithm is a well-known technology and will not be elaborated here. The bandwidth of the mean shift algorithm is set to 6. As other implementation manners, the implementer can set it according to the actual situation.

[0071] It should be noted that for the convenience of understanding, assume that the original position coordinates of a certain extreme value pixel are (x 0 , y 0 ). After one iteration by the mean shift algorithm, the new position coordinates are (x 1 , y 1 ). Then the drift direction angle of this iteration is

[0072]

[0073] The mean of the drift direction angles of all sub-iterations of all extreme value pixels on each edge contour in the surface gray-scale image of each azimuth is used as the drift trend degree of each edge contour in the surface gray-scale image of each azimuth.

[0074] It should be noted that the larger the drift trend degree, the greater the difference in the drift direction of the extreme value pixels after each iteration, and the greater the possibility that the corresponding extreme value pixels are on the edge contour. On the contrary, the smaller the drift trend degree, the more consistent the drift direction of the extreme value pixels after each iteration, and the greater the possibility that the corresponding extreme value pixels are on the crack contour.

[0075] Count the number of all extreme value pixels on each edge contour, denoted as the first quantity.

[0076] Count the number of all edge pixels on each edge contour, denoted as the second quantity.

[0077] Calculate the ratio of the first quantity to the second quantity, and take the product of the reciprocal of the drift trend degree and the ratio as the shape feature value of each edge contour in the surface gray-scale image of each azimuth.

[0078] Preferably, in this embodiment, the calculation method of the shape feature value of each edge contour in the surface gray-scale image of each azimuth is as follows: Among them, Z m,k is the shape feature value of the k-th edge contour in the surface gray-scale image of the m-th azimuth, n m,k is the number of all extreme value pixels on the k-th edge contour in the surface gray-scale image of the m-th azimuth, that is, the first quantity, and N m,kis the number of all edge pixel points on the k-th edge contour in the surface gray-scale image in the m-th azimuth, that is, the second quantity. is the drift trend degree of the k-th edge contour in the surface gray-scale image in the m-th azimuth.

[0079] It should be noted that the larger the ratio, the more extreme pixel points there are in the corresponding edge contour, indicating that the corresponding edge contour is more likely to be a crack contour; secondly, the smaller the drift trend degree, the more consistent the drift direction of the extreme pixel points after each iteration, so the probability of the corresponding extreme pixel points on the crack contour is greater, and the larger the obtained shape feature value, the greater the probability that the edge contour is a crack.

[0080] Thus, the shape feature values of each edge contour in the surface gray-scale image of each azimuth are obtained.

[0081] Step 4: Based on the color feature value and the shape feature value, determine the discrimination coefficient of each edge contour in the surface gray-scale image of each azimuth, and based on the discrimination coefficients of all azimuths, detect the processing quality of the mining parts.

[0082] Furthermore, based on the color feature value and the shape feature value, determine the discrimination coefficient, specifically:

[0083] Take the normalized result of the product of the color feature value and the shape feature value as the discrimination coefficient of each edge contour in the surface gray-scale image of each azimuth.

[0084] Preferably, in this embodiment, the arctangent function is used for normalization processing. The arctangent function is a well-known technology and will not be elaborated here. In this embodiment, the calculation method of the crack discrimination coefficient of each edge contour in the surface gray-scale image of each azimuth is: where w m,k is the discrimination coefficient of the k-th edge contour in the surface gray-scale image in the m-th azimuth, C m,k is the color feature value of the k-th edge contour in the surface gray-scale image in the m-th azimuth, Z m,k is the shape feature value of the k-th edge contour in the surface gray-scale image in the m-th azimuth, and arctan() is the arctangent function.

[0085] It should be noted that the larger the discrimination coefficient, the more likely the corresponding edge contour is a crack contour.

[0086] Furthermore, based on the discrimination coefficient, detect the processing quality of the mining parts, specifically:

[0087] If there are edge contours with the discrimination coefficient greater than the preset threshold in the surface grayscale images in all directions, the mining component has defects; otherwise, the mining component has no defects.

[0088] Preferably, in this embodiment, the preset threshold is set to 0.85. As other implementation manners, the implementer can set it according to the actual situation.

[0089] The embodiment of the present application also provides a mining component processing quality detection device, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the steps of any one of the above-mentioned mining component processing quality detection methods are implemented.

[0090] Based on the same inventive concept as the above method, the embodiment of the present application also provides a mining component processing quality detection system. A computer program is stored in the system, and when the computer program is executed by a processor, the steps of any one of the above-mentioned mining component processing quality detection methods are implemented.

[0091] It should be understood that although Figure 1 the steps in the flowchart are shown in sequence according to the indication of the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, Figure 1 at least a part of the steps in

[0092] can include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same moment, but can be executed at different moments. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or sub-steps or stages of other steps.

[0093] The technical features of the above embodiments can be combined arbitrarily. For the sake of brief description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.

Claims

1. A method for detecting the processing quality of mining parts, characterized in that: The method comprises the following steps: Obtain the surface grayscale image of each orientation of the mining parts, and extract each edge contour in the surface grayscale image of each orientation; Determine the directional discreteness of each edge contour in the surface grayscale image of each orientation according to the discreteness of the gradient direction of the edge pixel points on each edge contour; determine the grayscale discreteness of each edge contour in the surface grayscale image of each orientation according to the discreteness of the distribution of the grayscale values ​​of the pixel points in the area surrounded by each edge contour; determine the color feature value of each edge contour in the surface grayscale image of each orientation according to the fluctuation range of the grayscale values ​​of the pixel points in the area surrounded by each edge contour, combined with the directional discreteness and the grayscale discreteness; According to the curvature characteristics of the edge pixels on each edge contour and the change in the drift direction of the edge pixels corresponding to the extreme values ​​of all curvatures after mean shift, the drift trend of each edge contour in the surface grayscale image of each orientation is determined; according to the proportion of the number of edge pixels corresponding to the extreme values ​​of all curvatures on each edge contour, combined with the drift trend, the shape characteristic value of each edge contour in the surface grayscale image of each orientation is determined; Based on the color feature values ​​and the shape feature values, the discrimination coefficient of each edge contour in the surface grayscale image of each orientation is determined, and based on the discrimination coefficients of all orientations, the processing quality of mining parts is detected.

2. A method for detecting the processing quality of mining parts as claimed in claim 1, characterized in that: The directional dispersion of each edge contour in the surface grayscale image of each orientation is the variance of the gradient direction angles of all edge pixel points on each edge contour.

3. A method for detecting the processing quality of mining parts as claimed in claim 1, characterized in that: Determining the grayscale discreteness of each edge contour in the surface grayscale image at each orientation includes: The pixels within the area surrounded by each edge contour are recorded as internal pixels; Calculate the average grayscale value of any internal pixel point of each edge contour and all pixels in its neighborhood, and record it as the local grayscale of any internal pixel point; The discrete degree of the local grayscale of all internal pixel points of each edge contour is used as the grayscale discrete degree of each edge contour in the surface grayscale image of each orientation.

4. A method for detecting the processing quality of mining parts as claimed in claim 3, characterized in that: The step of determining the color feature value of each edge contour in the surface grayscale image at each orientation includes: Calculate the local grayscale range of all internal pixel points of each edge contour; The product of the range, the grayscale dispersion and the directional dispersion is used as the color feature value of each edge contour in the surface grayscale image of each orientation.

5. A method for detecting the processing quality of mining parts as claimed in claim 1, characterized in that: Determining the drift tendency of each edge contour in the surface grayscale image at each orientation includes: Taking any edge pixel point on each edge contour as the starting point, all edge pixel points on each edge contour are arranged in a clockwise direction to obtain the extreme value of the curvature of all edge pixel points on each edge contour; All the extreme value corresponding edge pixels are recorded as extreme value pixels; the position coordinates of all extreme value pixels on each edge contour are used as the input of the mean shift algorithm to obtain the drift direction angle of each extreme value pixel in each iteration; The average of the drift direction angles of all extreme pixel points on each edge contour in the surface grayscale image of each orientation in all iterations is taken as the drift tendency degree of each edge contour in the surface grayscale image of each orientation.

6. A method for detecting the processing quality of mining parts as claimed in claim 5, characterized in that: The step of determining the shape feature value of each edge contour in the surface grayscale image at each orientation includes: Count the number of all extreme pixel points on each edge contour, and record it as the first number; Count the number of all edge pixels on each edge contour, and record it as the second number; The ratio of the first quantity to the second quantity is calculated, and the product of the inverse of the drift tendency degree and the ratio is used as the shape feature value of each edge contour in the surface grayscale image of each orientation.

7. A method for detecting the processing quality of mining parts as claimed in claim 1, characterized in that: The discrimination coefficient of each edge contour in the surface grayscale image of each orientation is a normalized result of the product of the color feature value and the shape feature value.

8. A method for detecting the processing quality of mining parts as claimed in claim 1, characterized in that: The detection of the processing quality of mining parts includes: if there is an edge contour with a discrimination coefficient greater than a preset threshold in the surface grayscale image of all directions, the mining parts have defects, otherwise, the mining parts do not have defects.

9. A mining parts processing quality inspection equipment, characterized in that: The device includes a memory, a processor, and a computer program stored in the memory and running on the processor, and is characterized in that when the processor executes the computer program, the steps of a mining parts processing quality detection method as described in any one of claims 1-8 are implemented.

Images