Detection method for electrical automatic feeding equipment

Abstract

The invention relates to the technical field of image data processing, in particular to a detection method for electrical automation feeding equipment, and the method comprises the steps: dividing a conveying belt into a plurality of initial conveying belt image blocks, and carrying out the merging according to the gray level distribution condition of adjacent pixel points between the initial conveying belt image blocks, further obtaining a corresponding contrast characteristic value according to the gray scale gradient distribution characteristic and the gray scale value distribution characteristic of the crack defect area in the surface defect combined image block, and obtaining a contrast weight according to the gray scale value difference change condition of the crack defect in the surface defect combined image block under different illumination intensity levels; according to the contrast characteristic value and the contrast weight, obtaining a crack characteristic inhibition degree representing non-saliency of the crack area, and according to the crack characteristic inhibition degree, improving an edge detection algorithm to complete defect detection of the conveyor belt. The detection accuracy of the surface crack defect of the conveying belt is higher.

Description

Technical Field

[0001] The present application relates to the technical field of image data processing, and in particular to a detection method for electrical automation feeding equipment. Background Art

[0002] Electrical automation feeding equipment uses transmission belts as the core conveying components, and selects appropriate connection components and control components, such as sensors, robotic arms, etc., that can perform automated operations according to material properties and production requirements. Commonly used electrical automation feeding equipment includes servo roller feeders, buffer feeders, feeders with integrated stirring and screening functions, and other types. Different types of electrical automation feeding equipment can transmit different materials, such as packaging, plastic products, hardware, etc.

[0003] During the operation of automated feeding equipment, high-intensity operation makes component safety testing extremely important. In particular, the damage to the transmission belt during material transmission. The transmission belt has the characteristics of easy installation, large carrying capacity, fast speed, and stability, and is widely used in the field of material transmission. However, during the high-intensity operation of the transmission belt, crack defects may appear on its surface. If the crack defects are not handled in time, the transmission belt may be torn, affecting the stability of the transmission belt and even potential safety hazards. The area corresponding to the crack defect is small and not easy to be observed by the naked eye. If the machine is stopped for detection, it will affect the overall transportation equipment. Therefore, it is necessary to use computer vision to perform real-time detection without stopping the machine.

[0004] However, since there may be dirty areas on the surface of the conveyor belt, and the dirty areas will not affect the operation of the conveyor belt, it is necessary to distinguish between dirty areas and crack defects when detecting cracks in the conveyor belt. The existing technology usually distinguishes between dirty areas and crack defects based on the grayscale value in the collected surface image, and uses the relatively random change of grayscale values ​​in the dirty area to distinguish between dirty areas and crack defects. However, considering the working environment of the conveyor belt, the grayscale values ​​of the dirty patch area and the crack area on the conveyor belt may be evenly distributed, resulting in low accuracy in detecting crack defects on the conveyor belt surface. Summary of the invention

[0005] In order to solve the technical problem of low accuracy in detecting crack defects on the surface of conveyor belts in the prior art, the purpose of this application is to provide a detection method for electrical automation feeding equipment. The technical solution adopted is as follows: The present application proposes a detection method for electrical automation feeding equipment, the method comprising: Acquire a conveyor belt surface image at each light intensity level, and divide the conveyor belt surface image into a preset number of initial conveyor belt image blocks; At each light intensity level, all initial conveyor belt image blocks are merged to obtain at least one surface defect merged image block according to the gray value distribution of adjacent pixels between adjacent initial conveyor belt image blocks; the contrast characteristic value of each surface defect merged image block at each light intensity level is obtained according to the change of the gray value of the pixel points in each surface defect merged image block along the running direction of the conveyor belt and the overall gray value difference distribution; According to the distribution of the gray value difference of the pixel points in the surface defect merged image block under different light intensity levels, the contrast weight of each surface defect merged image block under each light intensity level is obtained; according to the contrast characteristic value and the contrast weight, the crack feature suppression degree of each surface defect merged image block is obtained; Edge detection is performed on each surface defect merged image block according to the degree of crack feature suppression to obtain an improved edge detection result of each surface defect merged image block, and conveyor belt defect detection is performed based on the improved edge detection result.

[0006] Furthermore, the step of merging all initial conveyor belt image blocks to obtain at least one surface defect merged image block comprises: For any two adjacent initial conveyor belt image patches: One of the initial conveyor belt image blocks is used as the first initial conveyor belt image block, and the other initial conveyor belt image block is used as the second initial conveyor belt image block; the pixel points adjacent to the block boundary in the first initial conveyor belt image block are used as the first boundary pixel points; the pixel points adjacent to the block boundary in the second initial conveyor belt image block are used as the second boundary pixel points; the first boundary pixel points are matched with the second boundary pixel points to obtain at least two matching binary groups, and the first boundary pixel points in the matching binary groups are adjacent to the second boundary pixel points; the matching binary group in which the first boundary pixel points and the second boundary pixel points are both abnormal pixel points is used as an abnormal matching binary group; and the number of abnormal matching binary groups corresponding to the first initial conveyor belt image block and the second initial conveyor belt image block is counted; When the block boundary is parallel to the running direction of the conveyor belt, the first initial conveyor belt image block and the second initial conveyor belt image block whose number of abnormal matching tuples is greater than a preset first threshold are merged to obtain a corresponding surface defect merged image block; When the block boundary is not parallel to the running direction of the conveyor belt, the first initial conveyor belt image block and the second initial conveyor belt image block whose number of abnormal matching tuples is greater than a preset second threshold are merged to obtain a corresponding surface defect merged image block.

[0007] Furthermore, the method for obtaining the contrast characteristic value includes: Under any illumination intensity level, a surface defect merged image block is selected as a target surface defect merged image block; the initial conveyor belt image block contained in the target surface defect merged image block is used as a reference initial conveyor belt image block; the grayscale gradient mean of the pixel points in each reference initial conveyor belt image block is calculated along the running direction of the conveyor belt; the grayscale value variance corresponding to all the pixel points in each reference initial conveyor belt image block is calculated; The product of the maximum grayscale value variance corresponding to each reference initial conveyor belt image block and the corresponding grayscale gradient mean is taken as the average gradient amplitude corresponding to each reference initial conveyor belt image block; the mean of all average gradient amplitudes of the target surface defect merged image block is taken as the contrast feature value of the target surface defect merged image block.

[0008] Furthermore, the method for obtaining the contrast weight includes: In each surface defect merged image block at each light intensity level, the grayscale value mean of all abnormal pixels is taken as the first grayscale mean, the grayscale value mean of all other pixels except the abnormal pixels is taken as the second grayscale mean, and the difference between the second grayscale mean and the first grayscale mean is taken as the corresponding grayscale inter-class difference; The grayscale class differences corresponding to each surface defect merged image block under each light intensity level are maximized to obtain the contrast weight of each surface defect merged image block under each light intensity level.

[0009] Furthermore, the method for obtaining the degree of crack characteristic suppression includes: The product of the contrast feature value and the contrast weight is used as the grayscale feature intensity of each surface defect merged image block under each light intensity level; the cumulative sum of all contrast feature intensities corresponding to each abnormal merged image is used as the crack feature suppression degree of each surface defect merged image block.

[0010] Furthermore, the method for obtaining the improved edge detection result includes: In the process of processing each surface defect merged image block by the edge detection algorithm, the original grayscale gradient value of each pixel point after non-maximum suppression is multiplied by the normalized value of the negative correlation mapping value of the crack feature suppression degree of the surface defect merged image block where it is located, as the improved grayscale gradient value of each pixel point. According to the improved grayscale gradient value and the corresponding gradient direction, double threshold detection is performed to obtain the improved edge detection result corresponding to each surface defect merged image block.

[0011] Furthermore, the edge detection algorithm is a canny edge detection algorithm.

[0012] Further, the conveyor belt defect detection according to the improved edge detection result includes: When continuous edges appear in the improved edge detection result, the area corresponding to the continuous edges is taken as the crack defect area on the conveyor belt; when continuous edges do not appear in the improved edge detection result, there is no crack defect on the corresponding conveyor belt.

[0013] Furthermore, the method for acquiring abnormal pixels includes: The grayscale values ​​of the pixels in the conveyor belt surface image under each light intensity level are used to obtain the corresponding segmentation threshold through the maximum inter-class variance method; the pixels in the conveyor belt surface image under each light intensity level whose grayscale values ​​are less than the corresponding segmentation threshold are regarded as abnormal pixels.

[0014] Furthermore, the state warning of the electrical automation feeding equipment based on the defect detection result includes: When a crack defect area is detected, the alarm system sends out a warning signal, and uses the coordinates of the center point of the crack defect area as the positioning result, and synchronously sends the positioning result and the warning signal to the staff's terminal device.

[0015] This application has the following beneficial effects: Taking into account that the stress generated by the conveyor belt during movement is mainly along its movement direction, when cracks appear on the surface of the conveyor belt, the corresponding cracks will usually expand in the direction of movement, that is, the direction of the cracks on the surface of the conveyor belt is usually consistent with the running direction of the conveyor belt, so the grayscale gradient corresponding to the crack area along the running direction is small; and because the overall area of ​​the crack area is small and the distribution is relatively regular, the overall impact of the crack area on the grayscale value of each initial conveyor belt image block is relatively small; based on this feature, the embodiment of the present application combines the change of grayscale value along the running direction of the conveyor belt and the distribution of overall grayscale value differences to obtain the contrast eigenvalue of each surface defect merged image block, and by calculating the contrast eigenvalue, the subsequent distinction between cracks and dirt on the conveyor belt surface is more accurate, thereby improving the accuracy of detecting cracks on the conveyor belt surface. Considering that the grayscale values ​​corresponding to the crack area and the grayscale values ​​of the dirty area are affected differently under different light intensity levels, this application obtains the contrast weight of each surface defect merged image block according to the distribution of the grayscale value differences of the pixels in the surface defect merged image block under different light intensity levels. The contrast weight combines the different characteristics of the dirty area and the crack area under different light intensities, making the detection accuracy of the crack area on the surface of the conveyor belt higher. The edge detection algorithm is further improved according to the degree of crack feature suppression obtained by the contrast feature value and the contrast weight, completing the defect detection of cracks on the surface of the conveyor belt, and realizing non-stop detection during the operation of the electrical automation feeding equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application 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 application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0017] Figure 1 A flow chart of a detection method for electrical automation feeding equipment provided in one embodiment of the present application. DETAILED DESCRIPTION

[0018] In order to further explain the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the following is a detailed description of the detection method for an electrical automated feeding device proposed in the present application, its specific implementation method, structure, features and effects, 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 application belongs.

[0020] The specific scheme of the detection method for electrical automation feeding equipment provided by the present application is described in detail below with reference to the accompanying drawings.

[0021] See also Figure 1 , which shows a flow chart of a detection method for an electrical automation feeding device provided by an embodiment of the present application, the method comprising: Step S1: acquiring a conveyor belt surface image at each light intensity level, and dividing the conveyor belt surface image into a preset number of initial conveyor belt image blocks.

[0022] The present application utilizes industrial cameras and other equipment to collect surface data of a conveyor belt in a feeding device in real time, processes and analyzes the surface data to distinguish between oily areas and crack defects on the conveyor belt, extracts the crack defect areas from the surface data, sends the coordinate information of the crack defect areas to the operator, and issues a warning signal at the same time, thereby realizing real-time detection of electrical automation feeding equipment without stopping.

[0023] Specifically, considering the influence of the working environment of the electrical automation feeding equipment, such as coal mine transportation, dust workshop, etc., the light brightness of the surrounding environment of the conveyor belt will change, which will affect the detection results. Therefore, the embodiment of the present application first obtains the surface image of the conveyor belt under each light intensity level. The specific process is: during the operation of the conveyor belt, the camera is fixed and the shooting angle is adjusted, and a fixed light source is set around the camera so that the camera can clearly capture the surface of the conveyor belt, and as the conveyor belt runs, the captured video contains the complete surface of the conveyor belt, and further according to each video frame, the image stitching technology is used to obtain a complete surface image of the conveyor belt. Then, the light intensity of the light source is changed in turn to obtain the surface image of the conveyor belt under each light intensity level. It should be noted that the image stitching technology is a prior art well known to those skilled in the art and will not be further described here.

[0024] It should be noted that in the embodiment of the present application, the light intensity level is divided into three light intensity levels according to the lux size, including a weak light intensity level corresponding to 500 lux to 1000 lux, a medium light intensity level corresponding to 1000 lux to 2000 lux, and a strong light intensity level corresponding to 2000 lux to 10000 lux. In the embodiment of the present application, the surface image of the conveyor belt at the strong light intensity level of 4000 lux, the surface image of the conveyor belt at the medium light intensity level of 1500 lux, and the surface image of the conveyor belt at the weak light intensity level of 800 lux are obtained respectively. In one embodiment of the present application, the camera selects an industrial high-speed camera, and the light source adopts a fluorescent lamp with adjustable brightness.

[0025] It should be further explained that the implementer can select different numbers of light sources to increase the light intensity according to the implementation situation, but it is necessary to ensure that the light is evenly distributed on the surface image of the conveyor belt; and the implementation method of the light intensity level division and the specific numerical selection of the light intensity level can be determined by the implementer according to the specific implementation environment. This application does not impose any special restrictions on this.

[0026] It should be noted that since the subsequent analysis process is based on the grayscale value of each pixel in the conveyor belt surface image, the embodiment of the present application grayscales the conveyor belt surface image to obtain the corresponding grayscale image, and for the convenience of expression, all conveyor belt surface images in the subsequent process are grayscale images, and no further details will be given later.

[0027] Considering that when there is a defective area in the conveyor belt surface image, the corresponding defective area is usually very small, so that when analyzing the conveyor belt surface image, a large amount of unnecessary calculations are often generated. And when analyzing the conveyor belt surface image as a whole, due to the small area of ​​the defective area, the corresponding features are not obvious enough, which may cause the defective area to not be fully detected, further reducing the accuracy of defect detection. In order to avoid this situation, the embodiment of the present application divides the conveyor belt surface image into a preset number of initial conveyor belt image blocks. By dividing the conveyor belt surface image into blocks, it is possible to accurately detect each small block, and it is possible to avoid the analysis and calculation of a large number of normal image blocks according to the difference in gray value distribution between the defective area and the normal area, thereby reducing the amount of calculation. In the embodiment of the present application, the preset number of divisions is set to 100, the initial conveyor belt image blocks are the same in size and shape, and the boundary of the initial conveyor belt image block is parallel or perpendicular to the running direction of the conveyor belt. Considering that the conveyor belt surface image is usually a rectangular area, the embodiment of the present application divides the conveyor belt surface image into 100 rectangular initial conveyor belt image blocks of the same shape and size. It should be noted that the implementer can adjust the size of the preset number of divisions and the shape and size of the initial conveyor belt image block according to the specific implementation environment, which will not be further elaborated here.

[0028] Step S2: At each light intensity level, based on the grayscale value distribution of adjacent pixels between adjacent initial conveyor belt image blocks, all initial conveyor belt image blocks are merged to obtain at least one surface defect merged image block; based on the change of the grayscale value of the pixel points in each surface defect merged image block along the running direction of the conveyor belt and the overall grayscale value difference distribution, the contrast characteristic value of each surface defect merged image block at each light intensity level is obtained.

[0029] Considering that the embodiment of the present application does not divide the conveyor belt surface image based on the characteristics of the conveyor belt surface image itself, but only divides the conveyor belt surface image into multiple initial conveyor belt image blocks, when a defective area appears in the conveyor belt surface image, part of the defective area may be divided into multiple adjacent initial conveyor belt image blocks. Therefore, in order to accurately analyze each complete defective area, it is necessary to merge the multiple adjacent initial conveyor belt image blocks divided from the defective area. At each light intensity level, the embodiment of the present application merges all the initial conveyor belt image blocks according to the gray value distribution of the adjacent pixels between the adjacent initial conveyor belt image blocks to obtain at least one surface defect merged image block.

[0030] Preferably, merging all initial conveyor belt image blocks to obtain at least one surface defect merged image block comprises: For any two adjacent initial conveyor belt image blocks: take one of the initial conveyor belt image blocks as the first initial conveyor belt image block, and take the other initial conveyor belt image block as the second initial conveyor belt image block; take the pixel points adjacent to the block boundary in the first initial conveyor belt image block as the first boundary pixel points; take the pixel points adjacent to the block boundary in the second initial conveyor belt image block as the second boundary pixel points; match the first boundary pixel points with the second boundary pixel points to obtain at least two matching binary groups, and the first boundary pixel points in the matching binary group are adjacent to the second boundary pixel points. In an embodiment of the present application, the dividing line between the first initial conveyor belt image block and the second initial conveyor belt image block, that is, the block boundary, is a virtual straight line, that is, the straight line is not composed of pixels, but only divides the boundary between the first initial conveyor belt image block and the second initial conveyor belt image block, so the first boundary pixel points and the second boundary pixel points are adjacent along the block boundary, that is, the first boundary pixel points in the matching binary group are adjacent to the second boundary pixel points. Since the initial conveyor belt image blocks in the embodiment of the present application are rectangles of the same shape and size, the block boundaries are straight lines, and the line segment between the first boundary pixel point and the second boundary pixel point in each corresponding matching binary group is perpendicular to the straight line corresponding to the block boundary.

[0031] The matching tuples in which both the first boundary pixel point and the second boundary pixel point are abnormal pixel points are taken as abnormal matching tuples; the number of abnormal matching tuples corresponding to the first initial conveyor belt image block and the second initial conveyor belt image block is counted. The abnormal pixel points are the pixel points corresponding to the defective area. When the defective area is divided into two adjacent initial conveyor belt image blocks, the defective area will be truncated. There should be a certain number of defective area pixels on both sides of the corresponding truncated boundary, that is, when the defective area is divided into the first initial conveyor belt image block and the second initial conveyor belt image block, there should be a certain number of abnormal matching tuples between the corresponding first initial conveyor belt image block and the second initial conveyor belt image block. However, considering that the conveyor belt surface image may be affected by other external factors such as noise, so that some abnormal matching tuples are not related to the defective area, it is necessary to further determine the number of abnormal matching tuples to avoid the influence of external factors such as noise.

[0032] Preferably, the method for acquiring abnormal pixel points includes: The grayscale values ​​of the pixels in the conveyor belt surface image at each light intensity level are used through the maximum inter-class variance method to obtain the corresponding segmentation threshold; the pixels in the conveyor belt surface image at each light intensity level whose grayscale values ​​are less than the corresponding segmentation threshold are regarded as abnormal pixels. Considering that for the crack defects on the conveyor belt that need to be detected in the embodiment of the present application, the grayscale values ​​of the pixels in the corresponding area are always significantly smaller than the pixels in other normal areas, the defective areas with small grayscale values ​​can be clearly divided through threshold segmentation. However, considering that the grayscale values ​​of the pixels in some dirty areas are also always significantly smaller than the grayscale values ​​of other normal areas, the crack areas required by the embodiment of the present application cannot be screened out only by the maximum inter-class variance method, so further analysis is needed on this basis.

[0033] When the block boundary is parallel to the running direction of the conveyor belt, the first initial conveyor belt image block and the second initial conveyor belt image block, in which the number of abnormal matching tuples is greater than the preset first threshold, are merged to obtain a corresponding surface defect merged image block. In the embodiment of the present application, the preset first threshold is set to 5. Since the purpose of the embodiment of the present application is to detect defects in the crack area, the initial conveyor belt image blocks corresponding to the crack area should be merged as much as possible according to the characteristics of the crack. When the boundary between the first initial conveyor belt image block and the second initial conveyor belt image block is parallel to the running direction of the conveyor belt, considering that the extension direction of the crack is also usually parallel to the running direction of the conveyor belt, when the crack area is divided into the first initial conveyor belt image block and the second initial conveyor belt image block, it is usually divided along the midline of the width corresponding to the crack area, so the number of corresponding abnormal matching tuples is relatively large, and the corresponding preset first threshold value is relatively large.

[0034] When the block boundary is not parallel to the running direction of the conveyor belt, the first initial conveyor belt image block and the second initial conveyor belt image block, whose number of abnormal matching tuples is greater than the preset second threshold, are merged to obtain the corresponding surface defect merged image block. In the embodiment of the present application, the preset second threshold is set to 2. When the boundary between the first initial conveyor belt image block and the second initial conveyor belt image block is perpendicular to the running direction of the conveyor belt, if the crack area is divided into the first initial conveyor belt image block and the second initial conveyor belt image block, the crack area is usually truncated along the length direction, and the number of corresponding abnormal matching tuples is usually less than or equal to the width of the crack area. The crack area is elongated and its width is generally small, so the number of corresponding abnormal matching tuples is relatively small, and the corresponding preset second threshold value is relatively small.

[0035] It should be noted that when an initial conveyor belt image block simultaneously satisfies the merging conditions with multiple initial conveyor belt image blocks, the initial conveyor belt image block and the corresponding multiple initial conveyor belt image blocks are merged into a surface defect merged image block, that is, there may be more than two initial conveyor belt image blocks in a surface defect merged image block.

[0036] At this point, the surface defect merged image block corresponding to each defect area is obtained. However, considering that there may be dirty areas and crack defects in the defect area, and the existence of dirty areas will not affect the safety and stability of the conveyor belt, there is no need to detect the dirty areas, while the existence of crack defects will affect the stability of the conveyor belt and pose certain safety hazards. Therefore, in order to further complete the defect detection of the crack area, it is necessary to further analyze the surface defect merged image block.

[0037] Considering that the grayscale values ​​of the pixels corresponding to the crack defect area and the dirty area are smaller than the grayscale values ​​of the pixels in the normal area, and the areas corresponding to the crack defect area and the dirty area are different, the crack defect area and the dirty area have different effects on the overall grayscale value of the initial conveyor belt image block. And because the extension direction of the crack defect area is parallel to the running direction of the conveyor belt, and the shape corresponding to the dirty area is irregular, the grayscale values ​​of the pixels in the crack defect area and the dirty area change differently along the running direction of the conveyor belt. The embodiment of the present application obtains the contrast characteristic value of each surface defect merged image block at each light intensity level based on the change of the grayscale value of the pixels in each surface defect merged image block along the running direction of the conveyor belt and the distribution of the overall grayscale value difference.

[0038] Preferably, the method for obtaining the contrast characteristic value includes: Under any illumination intensity level, a surface defect merged image block is selected as the target surface defect merged image block; the initial conveyor belt image block contained in the target surface defect merged image block is used as the reference initial conveyor belt image block; the grayscale gradient mean of the pixels in each reference initial conveyor belt image block is calculated along the running direction of the conveyor belt. Considering that there is a certain grayscale difference between the defective area corresponding to cracks and dirt and the normal area, each reference initial conveyor belt image block usually has a certain grayscale gradient feature. In each reference initial conveyor belt image block, since the overall extension direction of the crack is always consistent with the direction of the conveyor belt and is relatively slender, the number of pixels with grayscale differences along the running direction of the conveyor belt is usually small, so the corresponding grayscale gradient mean is usually small; and since the dirty area has an irregular shape, the number of pixels with grayscale differences along the running direction of the conveyor belt is usually much larger than the crack defect area, so the corresponding grayscale gradient mean is usually larger than the crack defect area. Therefore, the crack area can be better detected by calculating the grayscale gradient mean.

[0039] Calculate the variance of the grayscale values ​​corresponding to all pixels in each reference initial conveyor belt image block. Since the grayscale values ​​of pixels in the normal area are usually consistent, and the grayscale values ​​of pixels corresponding to the defective area are relatively small compared to the normal area, there is a certain variance in the grayscale values ​​of all pixels in each reference initial conveyor belt image block. The defect characteristics of the crack area are not obvious, and the corresponding defective area is usually small, that is, the number of pixels with grayscale differences from the pixels in the normal area is small, so the corresponding grayscale value variance is usually small. The dirty area has an irregular shape and the area corresponding to the dirty area is usually relatively large, that is, when there is a dirty area in the reference initial conveyor belt image block, the number of pixels with grayscale differences from the pixels in the normal area is large, so the corresponding grayscale value variance is larger than the crack defect area. Therefore, the defect detection of the crack area can be better performed by calculating the grayscale value variance.

[0040] It can be known that in the target surface defect merged image block, the larger the gray value variance in each reference initial conveyor belt image block, the larger the corresponding gray gradient mean along the running direction of the conveyor belt, and the less significant the crack characteristics of the corresponding defect area. The embodiment of the present application characterizes the insignificance of the crack characteristics of the defect area through the contrast characteristic value, that is, the smaller the contrast characteristic value, the more significant the crack characteristics of the defect area in the corresponding surface defect merged image block. The embodiment of the present application takes the product of the maximum gray value variance corresponding to each reference initial image block and the corresponding gray gradient mean as the average gradient amplitude corresponding to each reference initial conveyor belt image block; and takes the mean of all average gradient amplitudes of the target surface defect merged image block as the contrast characteristic value of the target surface defect merged image block. It should be noted that the implementer can choose other methods besides maximization to process the gray value variance according to the specific implementation environment, such as calculating the normalized value of the gray value variance, which will not be further elaborated here.

[0041] In the embodiment of the present application, the method for obtaining the contrast feature value of the target surface defect merged image block is expressed in the formula as follows: in, Merge the contrast feature values ​​of the image patches for the target surface defects, Merge the image blocks for the target surface defects. The gray value variance corresponding to the reference initial conveyor belt image block, The maximum value of the variance of the grayscale values ​​corresponding to all the reference initial conveyor belt image blocks corresponding to the target surface defect merged image block, Merge the image blocks for the target surface defects. The gray gradient mean of the reference initial conveyor belt image block along the running direction of the conveyor belt, is the number of reference initial conveyor belt image patches in the target surface defect merged image patch, Pick the function for the maximum value, Merge the image blocks for the target surface defects. The value after the variance of the gray value corresponding to the reference initial conveyor belt image block is maximized.

[0042] It should be noted that, since each initial conveyor belt image block corresponding to the target surface defect merged image block has abnormal pixels with grayscale values ​​different from those of pixels in the normal area, the maximum value of the corresponding grayscale value variance cannot be 0. Further, according to the method for obtaining the contrast feature value of the target surface defect merged image block, the contrast feature value of each surface defect merged image block under each light intensity level is obtained.

[0043] In addition, the implementer may also obtain the contrast characteristic value of the target surface defect merged image block through other forms of formulas, such as: in, is the normalization function, that is Merge the image blocks for the target surface defects. The normalized value of the grayscale value variance of the reference initial conveyor belt image block, and the meanings of the remaining parameters are the same as the formula corresponding to the method for obtaining the contrast feature value of the target surface defect merged image block in the embodiment of the present application, and will not be further elaborated here.

[0044] It should be noted that in the embodiments of the present application, the normalization method adopts linear normalization. The implementer may adopt other normalization methods according to the specific implementation environment. Linear normalization is an existing technology well known to those skilled in the art and will not be further described here.

[0045] Step S3: According to the distribution of the gray value difference of the pixels in the surface defect merged image block under different light intensity levels, the contrast weight of each surface defect merged image block under each light intensity level is obtained; according to the contrast eigenvalue and the contrast weight, the crack feature suppression degree of each surface defect merged image block is obtained.

[0046] Since the grayscale characteristics of the defect area corresponding to the same surface defect merged image block are different under different lighting conditions, analyzing the conveyor belt surface image under only one light intensity level may cause misjudgment of defect detection. Therefore, the embodiment of the present application obtains the contrast characteristic value of each surface defect merged image block under each light intensity level. Considering that the grayscale values ​​of the dirty area and the crack defect area will change when the light intensity changes, but there are certain differences in the changes in the grayscale values ​​of the dirty area and the crack defect area. The embodiment of the present application obtains the contrast weight of each surface defect merged image block under each light intensity level according to the distribution of the grayscale value difference of the pixel points in the surface defect merged image block under different light intensity levels. The crack defects on the surface of the conveyor belt are further detected by the contrast weight according to the characteristics of the difference in the grayscale value changes of the dirty area and the crack defect area under different light intensities.

[0047] Preferably, the method for obtaining the contrast weight includes: In each surface defect merged image block at each illumination intensity level, the grayscale value mean of all abnormal pixels is taken as the first grayscale mean, the grayscale value mean of all other pixels except the abnormal pixels is taken as the second grayscale mean, and the difference between the second grayscale mean and the first grayscale mean is taken as the corresponding grayscale inter-class difference. The grayscale inter-class difference corresponding to each surface defect merged image block at each illumination intensity level is maximized to obtain the contrast weight of each surface defect merged image block at each illumination intensity level.

[0048] Since the abnormal pixel points in the embodiment of the present application are the pixel points that are smaller than the corresponding segmentation threshold after being divided by the maximum inter-class variance method, that is, the pixel points corresponding to the defect area, all other pixel points other than the abnormal pixel points are normal pixel points. Therefore, the grayscale inter-class variance is the difference in the corresponding overall grayscale value between the normal pixel points and the abnormal pixel points at each light intensity level.

[0049] As the light intensity increases, the grayscale values ​​of the pixels on the conveyor belt surface will generally increase. However, since the crack defect area is different from the dirty area, the crack defect is a depression that passes through the conveyor belt. Although the grayscale value of the corresponding pixel will also increase with the increase of light intensity, compared with the dirty area and other normal pixel areas, the corresponding grayscale value increases at a slower rate, resulting in the greater the light intensity, the greater the difference between the grayscale classes corresponding to the crack defect area. The defect corresponding to the dirty area is attached to the surface of the conveyor belt. As the light intensity increases, the grayscale values ​​of the pixels in the dirty area and the pixels in the normal area increase at a similar rate, resulting in the grayscale class differences corresponding to the dirty area not changing much when the light intensity increases.

[0050] On the other hand, since the grayscale class differences corresponding to the dirty area do not change much, after maximizing the grayscale class differences of the corresponding surface defect merged image blocks under different light intensities, the contrast weights of the surface defect merged image blocks corresponding to the dirty area under different light intensity levels are numerically large and similar. However, the grayscale class differences of the surface defect merged image blocks corresponding to the crack defect area increase with the increase of light intensity, so after maximizing the grayscale class differences of the corresponding surface defect merged image blocks under different light intensities, the contrast weights of the surface defect merged image blocks corresponding to the crack defect area under different light intensity levels are positively correlated with the light intensity level, and there are certain differences in the contrast weights under different light intensity levels. Therefore, the crack area in the conveyor belt can be further detected by calculating the distribution difference of the contrast weights of each surface defect merged image block under different light intensity levels.

[0051] In the embodiment of the present application, Merged image blocks of surface defects at different light intensity levels The method for obtaining the contrast weight is expressed in the formula as follows: in, For the Merged image blocks of surface defects at different light intensity levels The contrast weight, For the Merged image blocks of surface defects at different light intensity levels The second grayscale mean, For the Merged image blocks of surface defects at different light intensity levels The first grayscale mean, Merge image patches for surface defects The second grayscale mean, Merge image patches for surface defects The first grayscale mean, Pick the function for the maximum value, For the Merged image blocks of surface defects at different light intensity levels The grayscale difference between classes, Merge patches for surface defects at all light intensity levels The corresponding maximum grayscale difference between classes.

[0052] It should be noted that, since all contrast weights corresponding to the surface defect merged image block of the crack area are smaller than those of the dirty area, the larger the overall contrast weights of the surface defect merged image block, the less significant the corresponding crack defect feature. Merged image blocks of surface defects at different light intensity levels The contrast weight acquisition method of the surface defect merged image block is used to obtain the contrast weight of each surface defect merged image block under each light intensity level. It should be noted that since there must be abnormal pixels and normal pixels with different grayscale values ​​in the surface defect merged image block, the corresponding maximum value of the grayscale inter-class difference cannot be 0.

[0053] At this point, the contrast weight and contrast eigenvalue of each surface defect merged image block at each light intensity level that characterizes the significant features of the crack area are obtained. Therefore, the crack feature suppression degree of each surface defect merged image block can be further obtained based on the contrast eigenvalue and contrast weight. The crack feature suppression degree characterizes the insignificance of the crack features of the surface defect merged image block, that is, the greater the crack feature suppression degree, the less significant the crack features of the corresponding surface defect merged image block.

[0054] Preferably, the method for obtaining the degree of crack characteristic suppression includes: The product of the contrast eigenvalue and the contrast weight is used as the contrast feature intensity of each surface defect merged image block at each illumination intensity level; the cumulative sum of all contrast feature intensities corresponding to each abnormal merged image is used as the crack feature suppression degree of each surface defect merged image block. For each abnormal merged image block, since it corresponds to a contrast weight and a contrast eigenvalue at each illumination intensity level, and the difference between the dirty area and the crack defect in the contrast weight can only be analyzed as a whole, in order to better reflect the distinguishing features of the dirty area and the crack defect in the contrast weight, the embodiment of the present application multiplies the contrast weight and the contrast eigenvalue at each illumination intensity level, and then accumulates the corresponding product with other products, that is, all contrast feature intensities corresponding to each abnormal merged image block are accumulated, so that the obtained crack feature suppression degree can better combine the corresponding distinguishing features represented by the contrast weight and the contrast eigenvalue. That is, the smaller the corresponding contrast eigenvalue, the smaller the overall contrast weight, the smaller the corresponding crack feature suppression degree, and the more significant the crack feature of the corresponding surface defect merged image block, that is, the more likely there is a crack defect.

[0055] In the embodiment of the present application, the surface defect merged image block The method for obtaining the degree of crack characteristic suppression is expressed in the formula as follows: in, Merge image patches for surface defects The degree of crack characteristic suppression, For the Merged image blocks of surface defects at different light intensity levels The contrast weight, For the Merged image blocks of surface defects at different light intensity levels The contrast characteristic value of is the number of preset light intensity levels. In the implementation of the present application, the number of light intensity levels is 3, namely, a weak light intensity level of 800 lux, a medium light intensity level of 1500 lux, and a strong light intensity level of 4000 lux.

[0056] In addition, the implementer can also obtain the surface defect merged image block through other forms of formulas The crack feature suppression degree should be improved, but the surface defect merging image block should be guaranteed. The corresponding contrast eigenvalue is positively correlated with the degree of crack feature suppression, and the contrast weight is positively correlated with the degree of crack feature suppression. For example: The meaning of each parameter is the same as that of the surface defect merged image block in the embodiment of the present application. The method for obtaining the degree of crack characteristic suppression is the same as the corresponding formula, which will not be further elaborated here.

[0057] Step S4: performing edge detection on each surface defect merged image block according to the crack feature suppression degree, obtaining an improved edge detection result of each surface defect merged image block, and performing conveyor belt defect detection according to the improved edge detection result.

[0058] At this point, the crack feature suppression degree of each surface defect merged image block in the conveyor belt surface image is obtained, and the crack feature suppression degree characterizes the insignificance of the crack area. Therefore, the edge texture features of each surface defect merged image block can be further enhanced or suppressed according to the crack feature suppression degree, that is, the edge texture of the surface defect merged image block with strong crack feature significance is enhanced, and the edge texture of the surface defect merged image block with weak crack feature significance is weakened, so that the accuracy of detecting crack defects on the conveyor belt surface is higher. The embodiment of the present application performs edge detection on each surface defect merged image block by the crack feature suppression degree to obtain an improved edge detection result of each surface defect merged image block. Since the grayscale value of the crack defect corresponding to the embodiment of the present application is significantly different from that of the normal area, an edge detection algorithm is usually used for defect detection. Since the grayscale value of the dirty area is also significantly different from that of the normal area, the crack feature suppression degree is used to make the improved edge detection result more accurate for detecting crack defects.

[0059] Preferably, the method for obtaining the improved edge detection result includes: In the process of processing each surface defect merged image block by the canny edge detection algorithm, the original grayscale gradient value of each pixel after non-maximum suppression and the negative correlation mapping value of the crack feature suppression degree of the surface defect merged image block where it is located are multiplied as the improved grayscale gradient value of each pixel. According to the improved grayscale gradient value and the corresponding gradient direction, the improved edge detection result corresponding to each surface defect merged image block is obtained by double threshold detection. When the crack feature suppression degree is greater, the crack feature of the corresponding surface defect merged image block is less significant, so it is necessary to weaken the edge texture feature of the surface defect merged image block; conversely, when the crack feature suppression degree is smaller, the crack feature of the corresponding surface defect merged image block is more significant, so it is necessary to enhance the edge texture feature of the surface defect merged image block or weaken it less relative to the dirty area. Since the size of the grayscale gradient value can represent the obvious degree of edge texture features, the crack feature suppression degree of each surface defect merged image block is negatively correlated and multiplied with the corresponding grayscale gradient value to obtain the improved grayscale gradient value.

[0060] It should be noted that the specific process of the canny edge detection algorithm mainly includes Gaussian filtering, image gradient calculation, non-maximum suppression and dual threshold detection. The embodiment of the present application only improves the grayscale gradient value of each pixel after non-maximum suppression, and the remaining processes are all existing technologies well known to technical personnel in this field and will not be further elaborated here.

[0061] It should be further explained that the light intensity level corresponding to the surface defect merged image block for edge detection in the embodiment of the present application is set to a medium light intensity level of 1500 lux. Since the size of the light intensity level does not affect the position of abnormal pixels in the surface defect merged image block, the implementer can choose surface defect merged image blocks under other light intensity levels for edge detection according to the specific implementation environment, which will not be further elaborated here.

[0062] In the embodiment of the present application, the surface defect merged image block Medium pixel The corresponding improved gray gradient value acquisition method is expressed in the formula as follows: in, Merge image patches for surface defects Medium pixel The corresponding improved gray gradient value, Merge image patches for surface defects Medium pixel The corresponding original grayscale gradient value, Merge image patches for surface defects The corresponding crack feature suppression degree, is an exponential function with the natural constant e as base, It is a normalization function. In the embodiment of the present application, the normalization function adopts linear normalization. Linear normalization is a conventional technical means and will not be further described here. Merge image patches for surface defects The corresponding negative correlation mapping value of the crack feature suppression degree.

[0063] Further merge image blocks based on surface defects Medium pixel The corresponding improved grayscale gradient value acquisition method obtains the improved grayscale gradient value of each pixel in each surface defect merged image block. Since the crack feature suppression degree of the surface defect merged image block corresponding to the dirty area is greater than that of the crack defect, the grayscale gradient value of the pixel in the surface defect merged image block corresponding to the dirty area is weakened to a higher degree. Therefore, by obtaining the improved edge detection result, the interference of the dirty area on the crack defect detection will be significantly reduced.

[0064] Further, defect detection can be completed based on the improved edge detection result. The embodiment of the present application performs conveyor belt defect detection based on the improved edge detection result.

[0065] Preferably, performing conveyor belt defect detection according to the improved edge detection result comprises: When continuous edges appear in the improved edge detection result, the area corresponding to the continuous edges is taken as the crack defect area on the conveyor belt; when continuous edges do not appear in the improved edge detection result, there is no crack defect on the corresponding conveyor belt. Since the edge detection result is improved by the crack feature suppression degree in the embodiment of the present application, the edge texture in the corresponding conveyor belt surface image is weakened, and only the surface defect merged image block with significantly high corresponding crack features is weakened to a lesser extent, so the continuous edges that appear in the improved edge detection result are usually the corresponding crack defect area.

[0066] Furthermore, when a crack defect area is detected, the alarm system sends out a warning signal, and uses the coordinates of the center point of the crack defect area as the positioning result, and synchronously sends the positioning result and the warning signal to the staff's terminal device, so that the staff can repair the crack defect in time, avoid unnecessary safety problems, and realize non-stop detection during the operation of the electrical automation feeding equipment.

[0067] It should be noted that the implementer can set up an alarm system according to the specific implementation environment. The alarm system can communicate with computer equipment, and the communication methods include but are not limited to network cable communication, Bluetooth communication, etc. The alarm system consists of a communication module, an audio and light warning device, such as a buzzer, a flash light, etc., a signal transmission module, and a power interface. Among them, the communication module is used to communicate with computer equipment, the audio and light warning device is used to send warning signals, and the signal transmission module is used to transmit and receive signals with the terminal equipment of the staff; the power interface is used to connect to the power supply.

[0068] In summary, after dividing the conveyor belt into multiple initial conveyor belt image blocks, the present application merges the initial conveyor belt image blocks according to the grayscale distribution of adjacent pixels between the initial conveyor belt image blocks, so that the surface defect merged image block can contain the complete defect area, and further obtains the corresponding contrast characteristic value according to the grayscale gradient distribution characteristics and grayscale value distribution characteristics of the crack defect area in the surface defect merged image block, and obtains the contrast weight according to the grayscale value difference change of the crack defect in the surface defect merged image block under different light intensity levels, and obtains the crack feature suppression degree that characterizes the insignificance of the crack area according to the contrast feature value and the contrast weight, and completes the conveyor belt defect detection according to the improved edge detection algorithm based on the crack feature suppression degree. The present application has higher accuracy in detecting crack defects on the conveyor belt surface.

[0069] It should be noted that the above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

[0070] 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 detection method for electrical automation feeding equipment, characterized in that: The method comprises: Acquire a conveyor belt surface image at each light intensity level, and divide the conveyor belt surface image into a preset number of initial conveyor belt image blocks; At each light intensity level, all initial conveyor belt image blocks are merged to obtain at least one surface defect merged image block according to the gray value distribution of adjacent pixels between adjacent initial conveyor belt image blocks; the contrast characteristic value of each surface defect merged image block at each light intensity level is obtained according to the change of the gray value of the pixel points in each surface defect merged image block along the running direction of the conveyor belt and the overall gray value difference distribution; According to the distribution of the gray value difference of the pixel points in the surface defect merged image block under different light intensity levels, the contrast weight of each surface defect merged image block under each light intensity level is obtained; according to the contrast characteristic value and the contrast weight, the crack feature suppression degree of each surface defect merged image block is obtained; Edge detection is performed on each surface defect merged image block according to the degree of crack feature suppression to obtain an improved edge detection result for each surface defect merged image block. Conveyor belt defect detection is performed based on the improved edge detection result, and status warning of the electrical automation feeding equipment is completed based on the defect detection result.

2. A detection method for electrical automation feeding equipment according to claim 1, characterized in that: The step of merging all initial conveyor belt image blocks to obtain at least one surface defect merged image block comprises: For any two adjacent initial conveyor belt image patches: One of the initial conveyor belt image blocks is used as the first initial conveyor belt image block, and the other initial conveyor belt image block is used as the second initial conveyor belt image block; the pixel points adjacent to the block boundary in the first initial conveyor belt image block are used as the first boundary pixel points; the pixel points adjacent to the block boundary in the second initial conveyor belt image block are used as the second boundary pixel points; the first boundary pixel points are matched with the second boundary pixel points to obtain at least two matching binary groups, and the first boundary pixel points in the matching binary groups are adjacent to the second boundary pixel points; the matching binary group in which the first boundary pixel points and the second boundary pixel points are both abnormal pixel points is used as an abnormal matching binary group; and the number of abnormal matching binary groups corresponding to the first initial conveyor belt image block and the second initial conveyor belt image block is counted; When the block boundary is parallel to the running direction of the conveyor belt, the first initial conveyor belt image block and the second initial conveyor belt image block whose number of abnormal matching tuples is greater than a preset first threshold are merged to obtain a corresponding surface defect merged image block; When the block boundary is not parallel to the running direction of the conveyor belt, the first initial conveyor belt image block and the second initial conveyor belt image block whose number of abnormal matching tuples is greater than a preset second threshold are merged to obtain a corresponding surface defect merged image block.

3. A detection method for electrical automation feeding equipment according to claim 1, characterized in that: The method for obtaining the contrast characteristic value comprises: Under any illumination intensity level, a surface defect merged image block is selected as a target surface defect merged image block; the initial conveyor belt image block contained in the target surface defect merged image block is used as a reference initial conveyor belt image block; the grayscale gradient mean of the pixel points in each reference initial conveyor belt image block is calculated along the running direction of the conveyor belt; the grayscale value variance corresponding to all the pixel points in each reference initial conveyor belt image block is calculated; The product of the maximum grayscale value variance corresponding to each reference initial conveyor belt image block and the corresponding grayscale gradient mean is taken as the average gradient amplitude corresponding to each reference initial conveyor belt image block; the mean of all average gradient amplitudes of the target surface defect merged image block is taken as the contrast feature value of the target surface defect merged image block.

4. A detection method for electrical automation feeding equipment according to claim 1, characterized in that: The method for obtaining the contrast weight includes: In each surface defect merged image block at each light intensity level, the grayscale value mean of all abnormal pixels is taken as the first grayscale mean, the grayscale value mean of all other pixels except the abnormal pixels is taken as the second grayscale mean, and the difference between the second grayscale mean and the first grayscale mean is taken as the corresponding grayscale inter-class difference; The grayscale class differences corresponding to each surface defect merged image block under each light intensity level are maximized to obtain the contrast weight of each surface defect merged image block under each light intensity level.

5. The detection method for electrical automation feeding equipment according to claim 1, characterized in that: The method for obtaining the degree of crack characteristic suppression includes: The product of the contrast feature value and the contrast weight is used as the grayscale feature intensity of each surface defect merged image block under each light intensity level; the cumulative sum of all contrast feature intensities corresponding to each abnormal merged image is used as the crack feature suppression degree of each surface defect merged image block.

6. A detection method for electrical automation feeding equipment according to claim 1, characterized in that: The method for obtaining the improved edge detection result comprises: In the process of processing each surface defect merged image block by the edge detection algorithm, the original grayscale gradient value of each pixel point after non-maximum suppression is multiplied by the normalized value of the negative correlation mapping value of the crack feature suppression degree of the surface defect merged image block where it is located, as the improved grayscale gradient value of each pixel point. According to the improved grayscale gradient value and the corresponding gradient direction, double threshold detection is performed to obtain the improved edge detection result corresponding to each surface defect merged image block.

7. A detection method for electrical automation feeding equipment according to claim 6, characterized in that: The edge detection algorithm is a canny edge detection algorithm.

8. The detection method for electrical automation feeding equipment according to claim 1, characterized in that: The conveyor belt defect detection according to the improved edge detection result comprises: When continuous edges appear in the improved edge detection result, the area corresponding to the continuous edges is taken as the crack defect area on the conveyor belt; when continuous edges do not appear in the improved edge detection result, there is no crack defect on the corresponding conveyor belt.

9. A detection method for electrical automation feeding equipment according to claim 2 or 4, characterized in that: The method for acquiring abnormal pixel points comprises: The grayscale values ​​of the pixels in the conveyor belt surface image under each light intensity level are used to obtain the corresponding segmentation threshold through the maximum inter-class variance method; the pixels in the conveyor belt surface image under each light intensity level whose grayscale values ​​are less than the corresponding segmentation threshold are regarded as abnormal pixels.

10. The detection method for electrical automation feeding equipment according to claim 1, characterized in that: The state warning of the electrical automation feeding equipment based on the defect detection result includes: When a crack defect area is detected, the alarm system sends out a warning signal, and uses the coordinates of the center point of the crack defect area as the positioning result, and synchronously sends the positioning result and the warning signal to the staff's terminal device.

Images