WO2023134789A1 - Automatic inspection method for belt-type conveying device - Google Patents

Automatic inspection method for belt-type conveying device Download PDF

Info

Publication number
WO2023134789A1
WO2023134789A1 PCT/CN2023/085832 CN2023085832W WO2023134789A1 WO 2023134789 A1 WO2023134789 A1 WO 2023134789A1 CN 2023085832 W CN2023085832 W CN 2023085832W WO 2023134789 A1 WO2023134789 A1 WO 2023134789A1
Authority
WO
WIPO (PCT)
Prior art keywords
defect area
edge
suspected defect
sequence
pixels
Prior art date
Application number
PCT/CN2023/085832
Other languages
French (fr)
Chinese (zh)
Inventor
周军
Original Assignee
苏州德斯米尔智能科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 苏州德斯米尔智能科技有限公司 filed Critical 苏州德斯米尔智能科技有限公司
Publication of WO2023134789A1 publication Critical patent/WO2023134789A1/en

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0004—Industrial image inspection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/13—Edge detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/136—Segmentation; Edge detection involving thresholding
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10016—Video; Image sequence
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20048—Transform domain processing
    • G06T2207/20061—Hough transform
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30108—Industrial image inspection
    • G06T2207/30164—Workpiece; Machine component
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Definitions

  • the invention relates to the technical field of image processing, in particular to an automatic detection method for belt conveying equipment.
  • Conveyor belt is widely used in industrial production. It is a mechanized and automatic conveying tool for material handling system. It has the characteristics of efficient transportation and easy use, which can greatly save labor costs. Friction causes excessive tension on the conveyor belt, and it is in a state of heavy load for a long time, resulting in abnormal conditions such as cracks on the surface of the conveyor belt. In severe cases, the conveyor belt will be pulled horizontally and affect safe production. .
  • the existing technology is to determine the Hough transform threshold corresponding to each suspected crack connected domain according to the acquisition reliability value of the local area image of the conveyor belt and the number of tear pixels corresponding to each suspected crack connected domain, and then determine whether each suspected crack connected domain is is the crack area, so as to judge whether there are cracks in the conveyor belt.
  • This method obtains the suspected defect area by segmenting the suspected defect area, and then performs Hough line detection on the suspected defect area to obtain the crack; but this method does not take into account the conveyor belt.
  • pollutants with characteristics similar to cracks will be formed on the surface, resulting in inaccurate identification of crack defects, resulting in inaccurate detection of conveyor belt cracks.
  • the invention provides an automatic detection method for belt conveying equipment to solve the problem of inaccurate detection of cracks in existing conveyor belts.
  • An automatic detection method for belt conveying equipment of the present invention adopts the following technical scheme:
  • S5. Determine whether the suspected defective region is a real defective region according to the expansion degree of the edge pixel points of each suspected defective region and the edge burr rate corresponding to the suspected defective region.
  • the feature quantity of each pixel is determined as follows:
  • the eigenvalue of each pixel is obtained.
  • D ij represents the feature distance between two adjacent pixel points i, j, W 1 and W 2 both represent the assigned weight, Represents the Euclidean distance between two pixels, Q 1 represents the characteristic quantity of pixel i in the Hessian matrix, Q 2 represents the characteristic quantity of pixel j in the Hessian matrix,
  • edge burr rate of the suspected defect area is determined as follows:
  • the edge burr rate of the suspected defect area is obtained by accumulating and averaging the feature distances.
  • grayscale sequence of multiple circles of pixels around the edge of the suspected defect area is determined as follows:
  • the mutation degree of the gray sequence is determined as follows:
  • the cumulative sum is averaged to obtain the mutation degree of the gray sequence.
  • the expansion degree of the edge pixels of the suspected defect area is determined as follows:
  • the cumulative sum is averaged to obtain the mutation degree of the gray sequence
  • the method for judging whether the suspected defect area is a real defect area is:
  • the degree of abnormality of the suspected defect area is obtained according to the expansion degree of the edge pixel points of each suspected defect area and the edge burr rate of the corresponding suspected defect area;
  • the suspected defective area is determined as follows:
  • segmentation threshold of the Otsu threshold segmentation method to segment the grayscale image, divide the grayscale image into multiple regions, obtain all regions smaller than the segmentation threshold, and use all regions smaller than the segmentation threshold as suspected defect regions.
  • the present invention obtains the suspected defect area by using the Otsu threshold segmentation method, narrows the scope of identifying crack defects, and judges the suspected defect area, improves the detection efficiency and ensures the accuracy of detection; Secondly, the present invention obtains the burr ratio and the expansion degree of the edge pixels in the suspected defect area, wherein, if there is a crack in the conveying process of the conveyor belt, the burr rate and the expansion degree of the crack area will be relatively large. Therefore, the combination of the burr rate and The degree of expansion judges whether the suspected defect area is a real defect area, making the final result more accurate.
  • Fig. 1 is the structural representation of the embodiment of a kind of automatic detection method for belt conveying equipment of the present invention
  • Fig. 2 is a schematic diagram of an image acquisition device in an embodiment of an automatic detection method for belt conveyor equipment according to the present invention.
  • An embodiment of an automatic detection method for belt conveying equipment of the present invention includes:
  • the specific steps to obtain the grayscale image of the surface of the conveyor belt when it is started are as follows: as shown in Figure 2, set up a camera directly above one side of the conveyor belt, and use the camera to collect multiple images of the conveyor belt surface, and the aspect ratio of the collected images is 4:3 , converting each conveyor belt surface image into a grayscale image to obtain multiple grayscale images, and the present invention takes one of the grayscale images as an example.
  • the specific steps to obtain the suspected defect area are: divide the grayscale image into N areas, and label each area.
  • the label sequence is ⁇ N 1 , N 2 ,...,N n ⁇ .
  • the purpose of labeling is to facilitate subsequent defects. location determination.
  • the Hessian matrix of each pixel and further obtain the two eigenvectors of the Hessian matrix And the corresponding eigenvalues ⁇ 1 , ⁇ 2 ; because in the Hessian matrix, the largest eigenvalue and its eigenvector represent the maximum curvature and curvature direction of the curve, and the smallest eigenvalue represents the minimum curvature and curvature direction of the curve, which can be Reflect the texture trend within the neighborhood of the pixel point; therefore, according to the two eigenvalues and eigenvectors of each pixel point as the feature quantity of the pixel point, the feature quantity of the pixel point can be expressed as
  • the specific steps for obtaining the characteristic distance between adjacent pixels on the edge of each suspected defect area are: using the feature quantity of each pixel to obtain the characteristic distance between adjacent pixels on the edge of each suspected defect area, specifically expressed The formula is:
  • the degree of change of the edge is represented according to the characteristic distance between adjacent edge pixels, because if the edge is more uneven, the distance between two adjacent pixels will be farther, so the eigenvalue in the Hessian matrix
  • the characteristic distance of edge similar points is a comprehensive characteristic value; the characteristic distance of two adjacent edge pixels is calculated according to the spatial distance of the pixel point and the characteristic distance in the Hessian matrix.
  • the characteristic distance D ij is larger, then Indicates that the larger the span between two adjacent pixels, the more uneven the edge here, the greater the possibility of burrs, and when the value of
  • the specific steps to obtain the edge burr rate of the suspected defect area are: to obtain the number of edge pixels of each suspected defect area, according to the characteristic distance between adjacent pixels on the edge of each suspected defect area and the corresponding suspected defect area The number of edge pixels is used to obtain the edge burr rate corresponding to the suspected defect area, and the specific expression is:
  • D ij represents the characteristic distance between pixel points i and j
  • n represents the number of edge pixels of the suspected defect area t
  • J t represents the edge burr rate of the t-th suspected defect area.
  • this formula calculates the average value of the feature distance between two points. If there are more burrs on the edge, then J t is larger, that is, the formula calculates the average value of two adjacent pixel points in the suspected defect area to represent the The burrs on the edge of the suspected defect area, so the value change of J t can reflect the burr rate of the edge.
  • the specific steps to obtain the grayscale sequence of the edge pixels of each suspected defect area and the multi-circle pixels around the edge are as follows: first obtain the grayscale sequence ⁇ pw 1 , pw 2 , ..., pw n of the edge pixels of the suspected defective pixel area ⁇ , and then the pixels of the s circle extending outward along the edge, the grayscale sequence is ⁇ pr 1 , pr 2 , ..., pr m ⁇ s .
  • s s ⁇ , d i represents the i-th
  • the logic of the above expression is: when the distance between pixels of the same gray level is 0, the conveyor belt will no longer expand, that is The pixels in this circle are cut-off pixels.
  • the gray-scale sequence composed of the edge pixels of each suspected defect area is the first ring gray-scale sequence, and there are s gray-scale sequences in total.
  • the specific steps for obtaining the mutation degree of the corresponding gray-scale sequence by using the gray-scale value of the pixel in each gray-scale sequence are: obtaining the gray-scale mean value of the pixel in each gray-scale sequence, and according to the gray value of the pixel in each gray-scale sequence Gray value, the gray value of each pixel to get the degree of mutation of each gray sequence, the specific expression is:
  • B s represents the mutation degree of the sth gray-scale sequence
  • q represents the number of pixels in the sth grayscale sequence.
  • the formula is transformed from the variance formula, and the average value of the difference between the gray value of the pixel point of the gray scale sequence and the gray scale mean value is used to reflect the degree of dispersion of the gray value in the gray scale sequence, that is, to represent each The degree of mutation of the grayscale sequence.
  • the specific steps to obtain the mutation degree of the corresponding gray-scale sequence by using the gray-scale value of the pixel in each gray-scale sequence are: according to the gray-scale value of the pixel in each gray-scale sequence, obtain the sudden change of the gray-scale value in each gray-scale sequence
  • the number of pixels D s according to the number D s of gray-scale mutation pixels in each gray-scale sequence and the mutation degree B s of each gray-scale sequence, the expansion degree of the edge pixels of the suspected defect area is obtained, and the specific expression
  • the formula is as follows:
  • D s represents the number of gray-scale mutation pixels in each gray-scale sequence
  • B s represents the mutation degree of the s-th gray-scale sequence
  • s represents the s-th gray-scale sequence
  • Kz represents the edge of the suspected defect area The degree of dilation of pixels.
  • the diffusion degree of the edge is obtained according to the number of pixels with gray-scale mutations and the degree of gray-scale mutation in each gray-scale sequence, because the more gray-scale mutation points appear on the edge, the greater the degree of gray-scale change, indicating that The edge pixels of this area have changed during the movement, so it is possible to judge the abnormality of this area.
  • the suspected defect area is more likely to be a real defect area.
  • S5. Determine whether the suspected defective region is a real defective region according to the expansion degree of the edge pixel points of each suspected defective region and the edge burr rate corresponding to the suspected defective region.
  • the abnormal degree of the suspected defect area is obtained, and the specific expression is as follows:
  • ⁇ indicates the abnormality degree of the tth suspected defect area
  • Kz indicates the expansion degree of the edge pixel of the tth suspected defect area
  • Jt indicates the edge burr rate of the tth suspected defect area.
  • the hyperbolic tangent function is used to normalize it according to the proportional logic relationship, that is, Jy,
  • the bigger Kz is, the bigger th(Jy) and th(Kz) are within 0-1
  • the Euclidean formula is used to integrate the two normalized characteristic parameter values to obtain the abnormal degree of suspected crack defects in this area.
  • the abnormal degree of the suspected defect area is obtained, and the abnormal degree threshold is set, which is set according to the specific situation.
  • the present invention does not give empirical reference values.
  • the suspected defect area is a real defect area, and the real defect area is mapped to the grayscale image to obtain the crack area on the conveyor belt.
  • the present invention obtains the suspected defect area by using the Otsu threshold segmentation method, narrows the scope of identifying crack defects, and judges the suspected defect area, improves the detection efficiency and ensures the accuracy of detection; Secondly, the present invention obtains the burr ratio and the expansion degree of the edge pixels in the suspected defect area, wherein, if there is a crack in the conveying process of the conveyor belt, the burr rate and the expansion degree of the crack area will be relatively large. Therefore, the combination of the burr rate and The degree of expansion judges whether the suspected defect area is a real defect area, making the final result more accurate.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Quality & Reliability (AREA)
  • Image Analysis (AREA)
  • Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
  • Image Processing (AREA)

Abstract

The present invention relates to the technical field of image processing, and relates in particular to an automatic inspection method for a belt-type conveying device, comprising: obtaining suspected defect areas; according to feature distances between adjacent pixel points on edges of each suspected defect area, obtaining edge burr rates corresponding to the suspected defect areas; obtaining a grayscale sequence of edge pixel points and edge peripheral multi-circle pixel points of each suspected defect area; using grayscale values of the pixel points in each grayscale sequence to obtain a mutation degree of the corresponding grayscale sequence; according to the mutation degree of each grayscale sequence and the number of grayscale mutation points, obtaining an expansion degree of the edge pixel points of each suspected defect area; and according to the expansion degree of the edge pixel points of each suspected defect area and the edge burr rate of the corresponding suspected defect area, determining whether the suspected defect area is a real defect area. The present invention increases the accuracy of conveying belt crack inspection.

Description

一种用于带式输送设备的自动检测方法An automatic detection method for belt conveying equipment 技术领域technical field
本发明涉及图像处理技术领域，具体涉及一种用于带式输送设备的自动检测方法。The invention relates to the technical field of image processing, in particular to an automatic detection method for belt conveying equipment.
背景技术Background technique
传送带在工业生产中应用极为广泛，是一种物料搬运系统机械化和自动化传送用具，具有高效运输，使用简易等特点，能极大的节省人工成本，但是在长时间的使用过程中，会因为物料摩擦使传送带受到的张力过大，长时间处于较大的负重状态，导致传送带的表面出现裂缝等异常情况，严重时会造成传送带横向拉断，影响安全生产，因此，需要对传送带的裂缝进行检测。Conveyor belt is widely used in industrial production. It is a mechanized and automatic conveying tool for material handling system. It has the characteristics of efficient transportation and easy use, which can greatly save labor costs. Friction causes excessive tension on the conveyor belt, and it is in a state of heavy load for a long time, resulting in abnormal conditions such as cracks on the surface of the conveyor belt. In severe cases, the conveyor belt will be pulled horizontally and affect safe production. .
现有技术是根据传送带局部区域图像的获取可靠程度值和各个疑似裂缝连通域所对应的撕裂像素个数，确定各个疑似裂缝连通域对应的霍夫变换阈值，进而确定各个疑似裂缝连通域是否为裂缝区域，从而判断传送带是否存在裂缝，该方法通过对疑似缺陷的区域的分割得到疑似缺陷的区域，再对疑似缺陷的区域进行霍夫直线检测得到裂缝；但是该方法没有考虑到传送带在长时间的使用过程中，表面会形成与裂缝特征较为相似的污染物，造成对裂缝缺陷的识别的不准确，从而导致传送带裂缝检测不准确。The existing technology is to determine the Hough transform threshold corresponding to each suspected crack connected domain according to the acquisition reliability value of the local area image of the conveyor belt and the number of tear pixels corresponding to each suspected crack connected domain, and then determine whether each suspected crack connected domain is is the crack area, so as to judge whether there are cracks in the conveyor belt. This method obtains the suspected defect area by segmenting the suspected defect area, and then performs Hough line detection on the suspected defect area to obtain the crack; but this method does not take into account the conveyor belt. During the use of time, pollutants with characteristics similar to cracks will be formed on the surface, resulting in inaccurate identification of crack defects, resulting in inaccurate detection of conveyor belt cracks.
发明内容Contents of the invention
本发明提供一种用于带式输送设备的自动检测方法，以解决现有的传送带裂缝检测不准确的问题。The invention provides an automatic detection method for belt conveying equipment to solve the problem of inaccurate detection of cracks in existing conveyor belts.
本发明的一种用于带式输送设备的自动检测方法，采用如下技术方案：An automatic detection method for belt conveying equipment of the present invention adopts the following technical scheme:
S1、获取传送带表面的灰度图像，根据Otsu阈值分割法对灰度图像进行分割得到疑似缺陷区域； S1. Obtain the grayscale image of the surface of the conveyor belt, and segment the grayscale image according to the Otsu threshold segmentation method to obtain suspected defect areas;
S2、获取每个疑似缺陷区域的边缘上每个像素点的灰度值的黑塞矩阵，获取每个黑塞矩阵对应的特征向量及特征值，根据每个黑塞矩阵对应的特征向量及特征值得到每个像素点的特征量；S2. Obtain the Hessian matrix of the gray value of each pixel on the edge of each suspected defect area, obtain the eigenvector and eigenvalue corresponding to each Hessian matrix, and obtain the eigenvector and eigenvalue corresponding to each Hessian matrix. Get the feature value of each pixel;
S3、根据每个像素点的特征量得到每个疑似缺陷区域的边缘上相邻像素点间的特征距离，根据每个疑似缺陷区域的边缘上相邻像素点间的特征距离和对应疑似缺陷区域的边缘像素点的个数得到对应疑似缺陷区域的边缘毛边率；S3. Obtain the characteristic distance between adjacent pixels on the edge of each suspected defect area according to the feature quantity of each pixel, and according to the characteristic distance between adjacent pixels on the edge of each suspected defect area and the corresponding suspected defect area The number of edge pixels of the corresponding to the edge burr rate of the suspected defect area;
S4、分别获取每个疑似缺陷区域边缘像素点及边缘外围多圈像素点的灰度序列，利用每个灰度序列得到每个疑似缺陷区域边缘像素点的扩张程度；S4. Obtain the grayscale sequence of the edge pixels of each suspected defect area and the multi-circle pixel points around the edge respectively, and use each grayscale sequence to obtain the expansion degree of the edge pixels of each suspected defect area;
S5、根据每个疑似缺陷区域边缘像素点的扩张程度及对应疑似缺陷区域的边缘毛边率判断该疑似缺陷区域是否为真实缺陷区域。S5. Determine whether the suspected defective region is a real defective region according to the expansion degree of the edge pixel points of each suspected defective region and the edge burr rate corresponding to the suspected defective region.
进一步的，所述每个像素点的特征量是按如下方法确定的：Further, the feature quantity of each pixel is determined as follows:
获取每个黑塞矩阵中两个对应的特征向量和特征值的乘积；Obtain the product of the two corresponding eigenvectors and eigenvalues in each Hessian matrix;
根据每个黑塞矩阵中两个对应的特征向量和特征值的乘积得到每个像素点的特征量。According to the product of two corresponding eigenvectors and eigenvalues in each Hessian matrix, the eigenvalue of each pixel is obtained.
进一步的，所述相邻像素点间的特征距离的具体表达式为：
Further, the specific expression of the characteristic distance between the adjacent pixel points is:
式中：Dij表示相邻两个像素点i，j之间的特征距离，W1、W2均表示分配的权重，表示两个像素点的欧式距离，Q1表示像素点i的在黑塞矩阵中的特征量，Q2表示像素点j的在黑塞矩阵中的特征量，|Q1-Q2|表示像素点i，j在黑塞矩阵中的特征距离。In the formula: D ij represents the feature distance between two adjacent pixel points i, j, W 1 and W 2 both represent the assigned weight, Represents the Euclidean distance between two pixels, Q 1 represents the characteristic quantity of pixel i in the Hessian matrix, Q 2 represents the characteristic quantity of pixel j in the Hessian matrix, |Q 1 -Q 2 | represents the pixel The characteristic distance of point i, j in the Hessian matrix.
进一步的，所述疑似缺陷区域的边缘毛边率是按如下方法确定的：Further, the edge burr rate of the suspected defect area is determined as follows:
对疑似缺陷区域边缘上相邻像素点之间的特征距离求和得到特征距离累加和；Summing the feature distances between adjacent pixels on the edge of the suspected defect area to obtain the cumulative sum of feature distances;
对特征距离累加和求平均得到该疑似缺陷区域的边缘毛边率。 The edge burr rate of the suspected defect area is obtained by accumulating and averaging the feature distances.
进一步的，所述疑似缺陷区域边缘外围多圈像素点的灰度序列是按如下方法确定的：Further, the grayscale sequence of multiple circles of pixels around the edge of the suspected defect area is determined as follows:
再沿着疑似缺陷像素点区域边缘向外延伸s圈，其中，s的截止条件应为最外圈相邻像素点之间的距离不再发生变化，获取最外侧圈层像素点的灰度序列后停止向外延伸圈层；Then extend the s circle outward along the edge of the suspected defective pixel area, where the cut-off condition of s should be that the distance between the adjacent pixels in the outermost circle does not change any more, and obtain the grayscale sequence of the pixels in the outermost circle Then stop extending the circle layer outward;
获取疑似缺陷区域边缘外围多圈像素点的灰度序列。Obtain the grayscale sequence of multiple circles of pixels around the edge of the suspected defect area.
进一步的，所述灰度序列的突变程度是按如下方法确定的：Further, the mutation degree of the gray sequence is determined as follows:
获取每个灰度序列中像素点的灰度均值和像素点个数；Obtain the gray mean and number of pixels in each gray sequence;
对每个灰度序列中每个像素点的灰度值与对应灰度序列中像素点的灰度均值的差值求和得到累加和；Summing the difference between the gray value of each pixel in each gray sequence and the gray mean value of the pixel in the corresponding gray sequence to obtain the cumulative sum;
对该累加和求平均得到该灰度序列的突变程度。The cumulative sum is averaged to obtain the mutation degree of the gray sequence.
进一步的，所述疑似缺陷区域边缘像素点的扩张程度是按如下方法确定的：Further, the expansion degree of the edge pixels of the suspected defect area is determined as follows:
获取每个灰度序列中像素点的灰度均值和像素点个数；Obtain the gray mean and number of pixels in each gray sequence;
对每个灰度序列中每个像素点的灰度值与对应灰度序列中像素点的灰度均值的差值求和得到累加和；Summing the difference between the gray value of each pixel in each gray sequence and the gray mean value of the pixel in the corresponding gray sequence to obtain the cumulative sum;
对该累加和求平均得到该灰度序列的突变程度；The cumulative sum is averaged to obtain the mutation degree of the gray sequence;
获取每个灰度序列中灰度值突变像素点的个数，根据每个灰度序列的突变程度和每个灰度序列中灰度值突变像素点的个数得到该疑似缺陷区域边缘像素点的扩张程度。Obtain the number of grayscale value mutation pixels in each grayscale sequence, and obtain the edge pixels of the suspected defect area according to the degree of mutation in each grayscale sequence and the number of grayscale value mutation pixels in each grayscale sequence degree of expansion.
进一步的，所述判断该疑似缺陷区域是否为真实缺陷区域的方法是：Further, the method for judging whether the suspected defect area is a real defect area is:
根据每个疑似缺陷区域边缘像素点的扩张程度及对应疑似缺陷区域的边缘毛边率得到该疑似缺陷区域的异常程度；The degree of abnormality of the suspected defect area is obtained according to the expansion degree of the edge pixel points of each suspected defect area and the edge burr rate of the corresponding suspected defect area;
设置异常程度阈值，当疑似缺陷区域的异常程度大于异常程度阈值该疑似缺陷区域为真实缺陷区域。Set the abnormal degree threshold, when the abnormal degree of the suspected defect area is greater than the abnormal degree threshold, the suspected defect area is the real defect area.
进一步的，所述疑似缺陷区域是按如下方法确定的： Further, the suspected defective area is determined as follows:
获取Otsu阈值分割法分割灰度图像的分割阈值，将灰度图像划分为多个区域，获取小于分割阈值的所有区域，将小于分割阈值的所有区域作为疑似缺陷区域。Obtain the segmentation threshold of the Otsu threshold segmentation method to segment the grayscale image, divide the grayscale image into multiple regions, obtain all regions smaller than the segmentation threshold, and use all regions smaller than the segmentation threshold as suspected defect regions.
本发明的有益效果是：本发明首先利用大津阈值分割法获取了疑似缺陷区域，缩小了识别裂缝缺陷的范围，对疑似缺陷区域进行判断，提高了检测效率的同时，保证了检测的准确性；其次，本发明获取了疑似缺陷区域的毛边率和边缘像素点的扩张程度，其中，传送带在传送过程中如果出现裂缝，则裂缝区域的毛边率和扩张程度都会较大，因此，结合毛边率和扩张程度判断疑似缺陷区域是否为真实缺陷区域，使最终得到的结果更加准确。The beneficial effects of the present invention are as follows: firstly, the present invention obtains the suspected defect area by using the Otsu threshold segmentation method, narrows the scope of identifying crack defects, and judges the suspected defect area, improves the detection efficiency and ensures the accuracy of detection; Secondly, the present invention obtains the burr ratio and the expansion degree of the edge pixels in the suspected defect area, wherein, if there is a crack in the conveying process of the conveyor belt, the burr rate and the expansion degree of the crack area will be relatively large. Therefore, the combination of the burr rate and The degree of expansion judges whether the suspected defect area is a real defect area, making the final result more accurate.
附图说明Description of drawings
为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动性的前提下，还可以根据这些附图获得其他的附图。In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only These are some embodiments of the present invention. For those skilled in the art, other drawings can also be obtained according to these drawings without any creative effort.
图1为本发明的一种用于带式输送设备的自动检测方法的实施例的结构示意图；Fig. 1 is the structural representation of the embodiment of a kind of automatic detection method for belt conveying equipment of the present invention;
图2为本发明的一种用于带式输送设备的自动检测方法的实施例中图像采集装置示意图。Fig. 2 is a schematic diagram of an image acquisition device in an embodiment of an automatic detection method for belt conveyor equipment according to the present invention.
具体实施方式Detailed ways
下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.
本发明的一种用于带式输送设备的自动检测方法的实施例，如图1所示，包括： An embodiment of an automatic detection method for belt conveying equipment of the present invention, as shown in Figure 1, includes:
S1、获取传送带启动时表面的灰度图像，获取Otsu阈值分割法分割灰度图像的分割阈值，将灰度图像划分为多个区域，获取小于分割阈值的所有区域，将小于分割阈值的所有区域作为疑似缺陷区域。S1. Obtain the grayscale image of the surface when the conveyor belt starts, obtain the segmentation threshold of the Otsu threshold segmentation method to segment the grayscale image, divide the grayscale image into multiple regions, obtain all regions smaller than the segmentation threshold, and divide all regions smaller than the segmentation threshold as a suspected defect area.
获取传送带启动时表面的灰度图像的具体步骤为：如图2所示，在传送带一侧的正上方设置相机，利用相机采集多张传送带表面图像，采集的图像的长宽比例为4:3，将每张传送带表面图像转化为灰度图像得到多张灰度图像，本发明以其中一张灰度图像为例。The specific steps to obtain the grayscale image of the surface of the conveyor belt when it is started are as follows: as shown in Figure 2, set up a camera directly above one side of the conveyor belt, and use the camera to collect multiple images of the conveyor belt surface, and the aspect ratio of the collected images is 4:3 , converting each conveyor belt surface image into a grayscale image to obtain multiple grayscale images, and the present invention takes one of the grayscale images as an example.
得到疑似缺陷区域的具体步骤为：将灰度图像均分为N个区域，对各个区域进行标号，标号序列为{N1,N2,…,Nn}，标号的目的是为了方便后续缺陷位置的确定。获取Otsu阈值分割法分割灰度图像的分割阈值，获取小于分割阈值的所有区域，将小于分割阈值的所有区域作为疑似缺陷区域。The specific steps to obtain the suspected defect area are: divide the grayscale image into N areas, and label each area. The label sequence is {N 1 , N 2 ,…,N n }. The purpose of labeling is to facilitate subsequent defects. location determination. Obtain the segmentation threshold of the Otsu threshold segmentation method to segment the gray image, obtain all regions smaller than the segmentation threshold, and use all regions smaller than the segmentation threshold as suspected defect regions.
S2、获取每个疑似缺陷区域的边缘上每个像素点的黑塞矩阵，获取每个黑塞矩阵对应的特征向量及特征值，根据每个黑塞矩阵对应的特征向量及特征值得到每个像素点的特征量。S2. Obtain the Hessian matrix of each pixel on the edge of each suspected defect area, obtain the eigenvector and eigenvalue corresponding to each Hessian matrix, and obtain each The feature quantity of the pixel point.
获得每一个像素点的黑塞矩阵，进一步求得黑塞矩阵的两个特征向量以及对应的特征值λ1，λ2；因为在黑塞矩阵中，最大的特征值及其特征向量表示曲线最大的曲率以及曲率方向，同样最小的特征值表示曲线最小的曲率以及曲率方向，可以反映该像素点邻域范围内的纹理走向；因此根据每一个像素点的两个特征值与特征向量作为该像素点的特征量，则像素点的特征量可表示为 Obtain the Hessian matrix of each pixel, and further obtain the two eigenvectors of the Hessian matrix And the corresponding eigenvalues λ 1 , λ 2 ; because in the Hessian matrix, the largest eigenvalue and its eigenvector represent the maximum curvature and curvature direction of the curve, and the smallest eigenvalue represents the minimum curvature and curvature direction of the curve, which can be Reflect the texture trend within the neighborhood of the pixel point; therefore, according to the two eigenvalues and eigenvectors of each pixel point as the feature quantity of the pixel point, the feature quantity of the pixel point can be expressed as
S3、根据每个像素点的特征量得到每个疑似缺陷区域的边缘上相邻像素点间的特征距离，根据每个疑似缺陷区域的边缘上相邻像素点间的特征距离和对应疑似缺陷区域的边缘像素点的个数得到对应疑似缺陷区域的边缘毛边率。S3. Obtain the characteristic distance between adjacent pixels on the edge of each suspected defect area according to the feature quantity of each pixel, and according to the characteristic distance between adjacent pixels on the edge of each suspected defect area and the corresponding suspected defect area The number of edge pixels of the corresponding to the edge burr rate of the suspected defect area.
得到每个疑似缺陷区域的边缘上相邻像素点间的特征距离的具体步骤为：利用每个像素点的特征量得到每个疑似缺陷区域的边缘上相邻像素点间的特征距离，具体表达式为：
The specific steps for obtaining the characteristic distance between adjacent pixels on the edge of each suspected defect area are: using the feature quantity of each pixel to obtain the characteristic distance between adjacent pixels on the edge of each suspected defect area, specifically expressed The formula is:
式中：Dij表示相邻两个像素点i，j之间的特征距离，W1、W2均表示分配的权重，W1＝0.7，W2＝0.3，表示两个像素点的欧式距离，Q1表示像素点i的在黑塞矩阵中的特征量，Q2表示像素点j的在黑塞矩阵中的特征量，|Q1-Q2|表示像素点i，j在黑塞矩阵中的特征距离。In the formula: D ij represents the feature distance between two adjacent pixel points i, j, W 1 and W 2 both represent the assigned weight, W 1 =0.7, W 2 =0.3, Represents the Euclidean distance between two pixels, Q 1 represents the characteristic quantity of pixel i in the Hessian matrix, Q 2 represents the characteristic quantity of pixel j in the Hessian matrix, |Q 1 -Q 2 | represents the pixel The characteristic distance of point i, j in the Hessian matrix.
其中，根据相邻边缘像素点之间的特征距离来表示边缘的变化程度，因为如果边缘越不平整，则相邻两个像素点之间的距离就会越远，因此黑塞矩阵中特征值与特征向量的乘积就会越大，边缘的曲率就会越大，并且根据边缘曲率可以判断边缘像素点的变化程度，因此通过像素点的欧式距离与黑塞矩阵中的特征距离来表示两个边缘相似点的特征距离，是一个综合特征值；根据像素点的空间距离与和黑塞矩阵中的特征距离计算得到相邻两个边缘像素点的特征距离，若特征距离Dij越大，则表明相邻两个像素点之间的跨度越大，则表示此处的边缘的越不平整，存在毛刺的可能越大，并且当|Q1-Q2|的值越大时，表示边缘曲线的曲率越大，进而也能得到边缘越不平整。Among them, the degree of change of the edge is represented according to the characteristic distance between adjacent edge pixels, because if the edge is more uneven, the distance between two adjacent pixels will be farther, so the eigenvalue in the Hessian matrix The greater the product with the eigenvector, the greater the curvature of the edge, and the degree of change of the edge pixels can be judged according to the edge curvature, so the Euclidean distance of the pixel and the characteristic distance in the Hessian matrix are used to represent two The characteristic distance of edge similar points is a comprehensive characteristic value; the characteristic distance of two adjacent edge pixels is calculated according to the spatial distance of the pixel point and the characteristic distance in the Hessian matrix. If the characteristic distance D ij is larger, then Indicates that the larger the span between two adjacent pixels, the more uneven the edge here, the greater the possibility of burrs, and when the value of |Q 1 -Q 2 | is larger, it indicates the edge curve The greater the curvature of , the more uneven the edge can be obtained.
得到疑似缺陷区域的边缘毛边率的具体步骤为：获取每个疑似缺陷区域的边缘像素点的个数，根据每个疑似缺陷区域的边缘上相邻像素点间的特征距离和对应疑似缺陷区域的边缘像素点的个数得到对应疑似缺陷区域的边缘毛边率，具体表达式为：
The specific steps to obtain the edge burr rate of the suspected defect area are: to obtain the number of edge pixels of each suspected defect area, according to the characteristic distance between adjacent pixels on the edge of each suspected defect area and the corresponding suspected defect area The number of edge pixels is used to obtain the edge burr rate corresponding to the suspected defect area, and the specific expression is:
式中：Dij表示像素点i，j之间的特征距离，n表示该疑似缺陷区域t的边缘像素点个数，Jt表示第t个疑似缺陷区域的边缘毛边率。In the formula: D ij represents the characteristic distance between pixel points i and j, n represents the number of edge pixels of the suspected defect area t, and J t represents the edge burr rate of the t-th suspected defect area.
其中，该公式是对两点之间的特征距离求平均值，如果边缘的毛刺较多，则Jt较大，即该公式对疑似缺陷区域中的两两相邻像素点求均值表征了该疑似缺陷区域边缘的毛刺情况，因此通过Jt的数值变化就能够反映边缘的毛边率。Among them, this formula calculates the average value of the feature distance between two points. If there are more burrs on the edge, then J t is larger, that is, the formula calculates the average value of two adjacent pixel points in the suspected defect area to represent the The burrs on the edge of the suspected defect area, so the value change of J t can reflect the burr rate of the edge.
S4、获取每个疑似缺陷区域边缘像素点及边缘外围多圈像素点的灰度序列，利用每个灰度序列中像素点的灰度值得到对应灰度序列的突变程度和灰度突变像素点个数，根据每个灰度序列的突变程度和灰度突变像素点个数得到每个疑似缺陷区域边缘像素点的扩张程度。 S4. Obtain the grayscale sequence of the edge pixels of each suspected defect area and the multi-circle pixels around the edge, and use the grayscale values of the pixels in each grayscale sequence to obtain the degree of mutation of the corresponding grayscale sequence and the grayscale mutation pixels. According to the mutation degree of each gray-scale sequence and the number of gray-scale mutation pixels, the expansion degree of the edge pixels of each suspected defect area is obtained.
获得疑似缺陷区域边缘的毛边率对该疑似缺陷区域进行判断会存在误差，因为脏污区域也会形成毛边，因此仅根据毛边率进行区分不够准确，因此在边缘毛边率的基础上，再进行污染物的特征区分。当传送带出现裂缝后，在传送带的运动过程中，受到张力的作用，裂缝会变大，但是污染物的大小不会随着传送带出现变化，所以通过计算疑似缺陷的边缘扩张程度，将污染物与裂纹缺陷进行区分。因为有裂缝的传送带在运动的过程中会受到张力的作用，随着运动的进行，裂缝会有逐渐增大的趋势，虽然在短时间的运动中其变化不是特别的明显，但是裂缝的边缘会出现拉丝状散开的线条，就相当于处于过拉状态后，横向的线条之间的空隙增大了。因此根据疑似裂缝缺陷处的扩张程度来判断是否裂痕缺陷。当出现拉丝状的线条时，原本纹理之间就会掺杂不属于该区域的像素点，并且根据先验知识，传送带表面像素点的灰度值比空隙间的像素点的灰度大，根据相邻像素点之间的灰度的变化来判断是否出现上述的拉丝状的线条。Obtaining the burr rate of the edge of the suspected defect area will cause errors in judging the suspected defect area, because the dirty area will also form burrs, so it is not accurate enough to distinguish only based on the burr rate, so on the basis of the edge burr rate, then carry out pollution Distinction of characteristics of objects. When the conveyor belt has cracks, the cracks will become larger under the action of tension during the movement of the conveyor belt, but the size of the pollutants will not change with the conveyor belt. Therefore, by calculating the degree of edge expansion of suspected defects, the pollutants and Crack defects are distinguished. Because the conveyor belt with cracks will be affected by tension during the movement, as the movement progresses, the cracks will gradually increase. Although the change is not particularly obvious during the short-term movement, the edges of the cracks will The appearance of brushed and scattered lines is equivalent to the increase of the gap between the horizontal lines after being in an over-drawn state. Therefore, it is judged whether there is a crack defect according to the expansion degree of the suspected crack defect. When brushed lines appear, pixels that do not belong to this area will be mixed between the original textures, and according to prior knowledge, the gray value of the pixels on the surface of the conveyor belt is larger than the gray value of the pixels in the gaps, according to The change of the gray level between adjacent pixels is used to determine whether the above-mentioned brushed lines appear.
获取每个疑似缺陷区域边缘像素点及边缘外围多圈像素点的灰度序列的具体步骤为：首先获得疑似缺陷像素点区域边缘像素点的灰度序列{pw1、pw2、…、pwn}，然后再沿着边缘向外延伸的s圈的像素点，其灰度序列为{pr1、pr2、…、prm}s。因为是描述像素点的扩张程度，因此s的截止条件应为最外圈相邻像素点之间的距离不再发生变化，获取最外侧圈层像素点的灰度序列后停止向外延伸圈层，因此计算同一圈的灰度相似的相邻像素点之间的距离，当距离为0时，s即截止，即当s＝{if di＝0|s＝s}，di表示第i圈相同灰度像素点之间的距离，si表示第i圈的像素点，上述表达式的逻辑为：当相同灰度的像素点之间的距离为0时，传送带不再出现扩张，即该圈的像素点为截止像素点。The specific steps to obtain the grayscale sequence of the edge pixels of each suspected defect area and the multi-circle pixels around the edge are as follows: first obtain the grayscale sequence {pw 1 , pw 2 , ..., pw n of the edge pixels of the suspected defective pixel area }, and then the pixels of the s circle extending outward along the edge, the grayscale sequence is {pr 1 , pr 2 , ..., pr m } s . Because it describes the expansion degree of pixels, the cut-off condition of s should be that the distance between adjacent pixels in the outermost circle no longer changes, and stop extending the circle layer outward after obtaining the grayscale sequence of pixels in the outermost circle layer , so calculate the distance between adjacent pixels with similar gray levels in the same circle, when the distance is 0, s is cut off, that is, when s={if d i =0|s=s}, d i represents the i-th The distance between pixels of the same gray level, s i represents the pixel of the i-th circle, the logic of the above expression is: when the distance between pixels of the same gray level is 0, the conveyor belt will no longer expand, that is The pixels in this circle are cut-off pixels.
需要说明的是，每个疑似缺陷区域的边缘像素点组成的灰度序列为第一圈灰度序列，共有s个灰度序列。It should be noted that the gray-scale sequence composed of the edge pixels of each suspected defect area is the first ring gray-scale sequence, and there are s gray-scale sequences in total.
利用每个灰度序列中像素点的灰度值得到对应灰度序列的突变程度的具体步骤为：获取每个灰度序列中像素点的灰度均值，根据每个灰度序列中像素点的灰度均值，每个像素点的灰度值得到每个灰度序列的突变程度，具体表达式为：
The specific steps for obtaining the mutation degree of the corresponding gray-scale sequence by using the gray-scale value of the pixel in each gray-scale sequence are: obtaining the gray-scale mean value of the pixel in each gray-scale sequence, and according to the gray value of the pixel in each gray-scale sequence Gray value, the gray value of each pixel to get the degree of mutation of each gray sequence, the specific expression is:
式中：Bs表示第s个灰度序列的突变程度,表示第s个灰度序列中第r个像素点的灰度值，表示第s灰度序列中像素点的灰度均值，q表示第s灰度序列中像素点的个数。In the formula: B s represents the mutation degree of the sth gray-scale sequence, Represents the gray value of the rth pixel in the sth grayscale sequence, Indicates the gray mean value of pixels in the sth grayscale sequence, and q represents the number of pixels in the sth grayscale sequence.
其中，该公式由方差公式转化而来，利用灰度序列像素点的灰度值与灰度均值的差值的平均值反映该该灰度序列中灰度值的离散程度，即表征了每个灰度序列的突变程度。Among them, the formula is transformed from the variance formula, and the average value of the difference between the gray value of the pixel point of the gray scale sequence and the gray scale mean value is used to reflect the degree of dispersion of the gray value in the gray scale sequence, that is, to represent each The degree of mutation of the grayscale sequence.
获得每个灰度序列中像素点的灰度值的突变个数，即获取每个灰度序列中相邻像素点灰度值的差值，设置差值阈值，根据具体情况设定，本发明不给出经验参考值，当相邻像素点灰度值的差值大于差值阈值时，相邻的两个像素点为突变像素点，灰度序列中每个像素点只统计一次，因此可得到每个灰度序列中灰度值突变像素点的个数。Obtain the number of sudden changes in the gray value of pixels in each gray sequence, that is, obtain the difference between the gray values of adjacent pixels in each gray sequence, and set the difference threshold, which is set according to specific conditions. The present invention No empirical reference value is given. When the difference between the gray values of adjacent pixels is greater than the difference threshold, the two adjacent pixels are mutation pixels, and each pixel in the gray sequence is counted only once, so it can be Obtain the number of gray value mutation pixels in each gray sequence.
利用每个灰度序列中像素点的灰度值得到对应灰度序列的突变程度的具体步骤为：根据每个灰度序列中像素点的灰度值获取每个灰度序列中灰度值突变像素点的个数Ds，根据每个灰度序列中灰度值突变像素点的个数Ds、每个灰度序列的突变程度Bs得到疑似缺陷区域边缘像素点的扩张程度，具体表达式如下：
The specific steps to obtain the mutation degree of the corresponding gray-scale sequence by using the gray-scale value of the pixel in each gray-scale sequence are: according to the gray-scale value of the pixel in each gray-scale sequence, obtain the sudden change of the gray-scale value in each gray-scale sequence The number of pixels D s , according to the number D s of gray-scale mutation pixels in each gray-scale sequence and the mutation degree B s of each gray-scale sequence, the expansion degree of the edge pixels of the suspected defect area is obtained, and the specific expression The formula is as follows:
式中：Ds表示每个灰度序列中灰度值突变像素点的个数，Bs表示第s个灰度序列的突变程度，s表示第s个灰度序列，Kz表示疑似缺陷区域边缘像素点的扩张程度。In the formula: D s represents the number of gray-scale mutation pixels in each gray-scale sequence, B s represents the mutation degree of the s-th gray-scale sequence, s represents the s-th gray-scale sequence, and Kz represents the edge of the suspected defect area The degree of dilation of pixels.
其中，根据每一条灰度序列中发生灰度突变的像素点的个数与灰度的突变程度得到边缘的扩散程度，因为边缘出现灰度突变点越多，灰度的变化程度越大，说明此区域的边缘像素点在运动的过程中出现了变化，因此能够判断此区域的异常，边缘出现灰度突变点越多，灰度的变化程度越大，则边缘像素点的扩张程度越大，该疑似缺陷区域越有可能是真实缺陷区域。 Among them, the diffusion degree of the edge is obtained according to the number of pixels with gray-scale mutations and the degree of gray-scale mutation in each gray-scale sequence, because the more gray-scale mutation points appear on the edge, the greater the degree of gray-scale change, indicating that The edge pixels of this area have changed during the movement, so it is possible to judge the abnormality of this area. The more gray-scale mutation points appear on the edge, the greater the degree of gray-scale change, and the greater the expansion of the edge pixels. The suspected defect area is more likely to be a real defect area.
S5、根据每个疑似缺陷区域边缘像素点的扩张程度及对应疑似缺陷区域的边缘毛边率判断该疑似缺陷区域是否为真实缺陷区域。S5. Determine whether the suspected defective region is a real defective region according to the expansion degree of the edge pixel points of each suspected defective region and the edge burr rate corresponding to the suspected defective region.
根据每个疑似缺陷区域边缘像素点的扩张程度及对应疑似缺陷区域的边缘毛边率得到该疑似缺陷区域的异常程度，具体表达式如下：
According to the expansion degree of the edge pixels of each suspected defect area and the edge burr rate of the corresponding suspected defect area, the abnormal degree of the suspected defect area is obtained, and the specific expression is as follows:
式中：ω表示第t个疑似缺陷区域的异常程度，Kz表示第t个疑似缺陷区域边缘像素点的扩张程度，Jt表示第t个疑似缺陷区域的边缘毛边率。In the formula: ω indicates the abnormality degree of the tth suspected defect area, Kz indicates the expansion degree of the edge pixel of the tth suspected defect area, and Jt indicates the edge burr rate of the tth suspected defect area.
其中，疑似缺陷区域边缘像素点的扩张程度及边缘毛边率两者的数值越大，疑似裂纹缺陷的异常程度越高，因此利用双曲正切函数将其按正比例逻辑关系归一化，即Jy、Kz越大，th(Jy)、th(Kz)在0-1内越大，则是利用欧几里得公式对两个归一化后的特征参数值进行整合，得到该区域疑似裂纹缺陷的异常程度。Among them, the greater the expansion degree of the edge pixels of the suspected defect area and the edge burr rate, the higher the abnormality of the suspected crack defect. Therefore, the hyperbolic tangent function is used to normalize it according to the proportional logic relationship, that is, Jy, The bigger Kz is, the bigger th(Jy) and th(Kz) are within 0-1, The Euclidean formula is used to integrate the two normalized characteristic parameter values to obtain the abnormal degree of suspected crack defects in this area.
根据每个疑似缺陷区域边缘像素点的扩张程度及对应疑似缺陷区域的边缘毛边率得到该疑似缺陷区域的异常程度，设置异常程度阈值，根据具体情况设定，本发明不给出经验参考值，当疑似缺陷区域的异常程度大于异常程度阈值该疑似缺陷区域为真实缺陷区域，将真是缺陷区域对应到灰度图像中得到传送带上的裂缝区域。According to the expansion degree of the edge pixel points of each suspected defect area and the edge burr rate of the corresponding suspected defect area, the abnormal degree of the suspected defect area is obtained, and the abnormal degree threshold is set, which is set according to the specific situation. The present invention does not give empirical reference values. When the abnormality of the suspected defect area is greater than the abnormality threshold, the suspected defect area is a real defect area, and the real defect area is mapped to the grayscale image to obtain the crack area on the conveyor belt.
本发明的有益效果是：本发明首先利用大津阈值分割法获取了疑似缺陷区域，缩小了识别裂缝缺陷的范围，对疑似缺陷区域进行判断，提高了检测效率的同时，保证了检测的准确性；其次，本发明获取了疑似缺陷区域的毛边率和边缘像素点的扩张程度，其中，传送带在传送过程中如果出现裂缝，则裂缝区域的毛边率和扩张程度都会较大，因此，结合毛边率和扩张程度判断疑似缺陷区域是否为真实缺陷区域，使最终得到的结果更加准确。The beneficial effects of the present invention are as follows: firstly, the present invention obtains the suspected defect area by using the Otsu threshold segmentation method, narrows the scope of identifying crack defects, and judges the suspected defect area, improves the detection efficiency and ensures the accuracy of detection; Secondly, the present invention obtains the burr ratio and the expansion degree of the edge pixels in the suspected defect area, wherein, if there is a crack in the conveying process of the conveyor belt, the burr rate and the expansion degree of the crack area will be relatively large. Therefore, the combination of the burr rate and The degree of expansion judges whether the suspected defect area is a real defect area, making the final result more accurate.
以上所述仅为本发明的较佳实施例而已，并不用以限制本发明，凡在本发明的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本发明的保护范围之内。 The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the scope of the present invention. within the scope of protection.

Claims (9)

  1. 一种用于带式输送设备的自动检测方法，其特征在于，包括：An automatic detection method for belt conveying equipment, characterized in that it comprises:
    S1、获取传送带表面的灰度图像，根据Otsu阈值分割法对灰度图像进行分割得到疑似缺陷区域；S1. Obtain the grayscale image of the surface of the conveyor belt, and segment the grayscale image according to the Otsu threshold segmentation method to obtain suspected defect areas;
    S2、获取每个疑似缺陷区域的边缘上每个像素点的灰度值的黑塞矩阵，获取每个黑塞矩阵对应的特征向量及特征值，根据每个黑塞矩阵对应的特征向量及特征值得到每个像素点的特征量；S2. Obtain the Hessian matrix of the gray value of each pixel on the edge of each suspected defect area, obtain the eigenvector and eigenvalue corresponding to each Hessian matrix, and obtain the eigenvector and eigenvalue corresponding to each Hessian matrix. Get the feature value of each pixel;
    S3、根据每个像素点的特征量得到每个疑似缺陷区域的边缘上相邻像素点间的特征距离，根据每个疑似缺陷区域的边缘上相邻像素点间的特征距离和对应疑似缺陷区域的边缘像素点的个数得到对应疑似缺陷区域的边缘毛边率；S3. Obtain the characteristic distance between adjacent pixels on the edge of each suspected defect area according to the feature quantity of each pixel, and according to the characteristic distance between adjacent pixels on the edge of each suspected defect area and the corresponding suspected defect area The number of edge pixels of the corresponding to the edge burr rate of the suspected defect area;
    S4、分别获取每个疑似缺陷区域边缘像素点及边缘外围多圈像素点的灰度序列，利用每个灰度序列得到每个疑似缺陷区域边缘像素点的扩张程度；S4. Obtain the grayscale sequence of the edge pixels of each suspected defect area and the multi-circle pixel points around the edge respectively, and use each grayscale sequence to obtain the expansion degree of the edge pixels of each suspected defect area;
    S5、根据每个疑似缺陷区域边缘像素点的扩张程度及对应疑似缺陷区域的边缘毛边率判断该疑似缺陷区域是否为真实缺陷区域。S5. Determine whether the suspected defective region is a real defective region according to the expansion degree of the edge pixel points of each suspected defective region and the edge burr rate corresponding to the suspected defective region.
  2. 根据权利要求1所述的一种用于带式输送设备的自动检测方法，其特征在于，所述每个像素点的特征量是按如下方法确定的：A kind of automatic detection method for belt conveying equipment according to claim 1, is characterized in that, the feature quantity of described each pixel is determined as follows:
    获取每个黑塞矩阵中两个对应的特征向量和特征值的乘积；Obtain the product of the two corresponding eigenvectors and eigenvalues in each Hessian matrix;
    根据每个黑塞矩阵中两个对应的特征向量和特征值的乘积得到每个像素点的特征量。According to the product of two corresponding eigenvectors and eigenvalues in each Hessian matrix, the eigenvalue of each pixel is obtained.
  3. 根据权利要求1所述的一种用于带式输送设备的自动检测方法，其特征在于，所述相邻像素点间的特征距离的具体表达式为：
    A kind of automatic detection method for belt conveying equipment according to claim 1, is characterized in that, the concrete expression of the characteristic distance between described adjacent pixel points is:
    式中：Dij表示相邻两个像素点i，j之间的特征距离，W1、W2均表示分配的权重，表示两个像素点的欧式距离，Q1表示像素点i的在黑塞矩阵中的特征量，Q2表示像素点j的在黑塞矩阵中的特征量，|Q1-Q2|表示像素点i，j在黑塞矩阵中的特征距离。 In the formula: D ij represents the feature distance between two adjacent pixel points i, j, W 1 and W 2 both represent the assigned weight, Represents the Euclidean distance between two pixels, Q 1 represents the characteristic quantity of pixel i in the Hessian matrix, Q 2 represents the characteristic quantity of pixel j in the Hessian matrix, |Q 1 -Q 2 | represents the pixel The characteristic distance of point i, j in the Hessian matrix.
  4. 根据权利要求1所述的一种用于带式输送设备的自动检测方法，其特征在于，所述疑似缺陷区域的边缘毛边率是按如下方法确定的：An automatic detection method for belt conveying equipment according to claim 1, wherein the edge burr rate of the suspected defect area is determined as follows:
    对疑似缺陷区域边缘上相邻像素点之间的特征距离求和得到特征距离累加和；Summing the feature distances between adjacent pixels on the edge of the suspected defect area to obtain the cumulative sum of feature distances;
    对特征距离累加和求平均得到该疑似缺陷区域的边缘毛边率。The edge burr rate of the suspected defect area is obtained by accumulating and averaging the feature distances.
  5. 根据权利要求1所述的一种用于带式输送设备的自动检测方法，其特征在于，所述疑似缺陷区域边缘外围多圈像素点的灰度序列是按如下方法确定的：An automatic detection method for belt conveying equipment according to claim 1, wherein the grayscale sequence of multiple circles of pixels on the periphery of the suspected defect area is determined as follows:
    再沿着疑似缺陷像素点区域边缘向外延伸s圈，其中，s的截止条件应为最外圈相邻像素点之间的距离不再发生变化，获取最外侧圈层像素点的灰度序列后停止向外延伸圈层；Then extend the s circle outward along the edge of the suspected defective pixel area, where the cut-off condition of s should be that the distance between the adjacent pixels in the outermost circle does not change any more, and obtain the grayscale sequence of the pixels in the outermost circle Then stop extending the circle layer outward;
    获取疑似缺陷区域边缘外围多圈像素点的灰度序列。Obtain the grayscale sequence of multiple circles of pixels around the edge of the suspected defect area.
  6. 根据权利要求1所述的一种用于带式输送设备的自动检测方法，其特征在于，所述灰度序列的突变程度是按如下方法确定的：A kind of automatic detection method for belt conveying equipment according to claim 1, is characterized in that, the mutation degree of described gray scale sequence is determined as follows:
    获取每个灰度序列中像素点的灰度均值和像素点个数；Obtain the gray mean and number of pixels in each gray sequence;
    对每个灰度序列中每个像素点的灰度值与对应灰度序列中像素点的灰度均值的差值求和得到累加和；Summing the difference between the gray value of each pixel in each gray sequence and the gray mean value of the pixel in the corresponding gray sequence to obtain the cumulative sum;
    对该累加和求平均得到该灰度序列的突变程度。The cumulative sum is averaged to obtain the mutation degree of the gray sequence.
  7. 根据权利要求1所述的一种用于带式输送设备的自动检测方法，其特征在于，所述疑似缺陷区域边缘像素点的扩张程度是按如下方法确定的：An automatic detection method for belt conveying equipment according to claim 1, wherein the expansion degree of the edge pixel points of the suspected defect area is determined as follows:
    获取每个灰度序列中像素点的灰度均值和像素点个数；Obtain the gray mean and number of pixels in each gray sequence;
    对每个灰度序列中每个像素点的灰度值与对应灰度序列中像素点的灰度均值的差值求和得到累加和；Summing the difference between the gray value of each pixel in each gray sequence and the gray mean value of the pixel in the corresponding gray sequence to obtain the cumulative sum;
    对该累加和求平均得到该灰度序列的突变程度；The cumulative sum is averaged to obtain the mutation degree of the gray sequence;
    获取每个灰度序列中灰度值突变像素点的个数，根据每个灰度序列的突变程度和每个灰度序列中灰度值突变像素点的个数得到该疑似缺陷区域边缘像素点的扩张程度。Obtain the number of grayscale value mutation pixels in each grayscale sequence, and obtain the edge pixels of the suspected defect area according to the degree of mutation in each grayscale sequence and the number of grayscale value mutation pixels in each grayscale sequence degree of expansion.
  8. 根据权利要求1所述的一种用于带式输送设备的自动检测方法，其特征在于，所述判断该疑似缺陷区域是否为真实缺陷区域的方法是： An automatic detection method for belt conveying equipment according to claim 1, wherein the method for judging whether the suspected defect area is a real defect area is:
    根据每个疑似缺陷区域边缘像素点的扩张程度及对应疑似缺陷区域的边缘毛边率得到该疑似缺陷区域的异常程度；The degree of abnormality of the suspected defect area is obtained according to the expansion degree of the edge pixel points of each suspected defect area and the edge burr rate of the corresponding suspected defect area;
    设置异常程度阈值，当疑似缺陷区域的异常程度大于异常程度阈值该疑似缺陷区域为真实缺陷区域。Set the abnormal degree threshold, when the abnormal degree of the suspected defect area is greater than the abnormal degree threshold, the suspected defect area is the real defect area.
  9. 根据权利要求1所述的一种用于带式输送设备的自动检测方法，其特征在于，所述疑似缺陷区域是按如下方法确定的：An automatic detection method for belt conveying equipment according to claim 1, wherein the suspected defect area is determined as follows:
    获取Otsu阈值分割法分割灰度图像的分割阈值，将灰度图像划分为多个区域，获取小于分割阈值的所有区域，将小于分割阈值的所有区域作为疑似缺陷区域。 Obtain the segmentation threshold of the Otsu threshold segmentation method to segment the grayscale image, divide the grayscale image into multiple regions, obtain all regions smaller than the segmentation threshold, and use all regions smaller than the segmentation threshold as suspected defect regions.
PCT/CN2023/085832 2022-10-25 2023-04-03 Automatic inspection method for belt-type conveying device WO2023134789A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202211308242.8 2022-10-25
CN202211308242.8A CN115375688B (en) 2022-10-25 2022-10-25 Automatic detection method for belt conveyor

Publications (1)

Publication Number Publication Date
WO2023134789A1 true WO2023134789A1 (en) 2023-07-20

Family

ID=84074296

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/085832 WO2023134789A1 (en) 2022-10-25 2023-04-03 Automatic inspection method for belt-type conveying device

Country Status (3)

Country Link
CN (1) CN115375688B (en)
LU (1) LU504264B1 (en)
WO (1) WO2023134789A1 (en)

Cited By (50)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116630815A (en) * 2023-07-25 2023-08-22 济南玖通志恒信息技术有限公司 Intelligent agricultural pest detection method
CN116630315A (en) * 2023-07-24 2023-08-22 山东东阿亿福缘阿胶制品有限公司 Intelligent beverage packaging defect detection method based on computer vision
CN116630329A (en) * 2023-07-26 2023-08-22 山东山森数控技术有限公司 Online visual detection method for multi-axis multi-channel numerical control system
CN116664569A (en) * 2023-07-31 2023-08-29 山东正华建筑科技有限公司 Weld flash defect detection method
CN116682107A (en) * 2023-08-03 2023-09-01 山东国宏生物科技有限公司 Soybean visual detection method based on image processing
CN116681695A (en) * 2023-07-27 2023-09-01 山东阁林板建材科技有限公司 Quality detection method for anti-deformation template end face
CN116721106A (en) * 2023-08-11 2023-09-08 山东明达圣昌铝业集团有限公司 Profile flaw visual detection method based on image processing
CN116735612A (en) * 2023-08-15 2023-09-12 山东精亿机械制造有限公司 Welding defect detection method for precise electronic components
CN116740057A (en) * 2023-08-11 2023-09-12 深圳市鹏基精密工业有限公司 Cylindrical workpiece burr online detection method and system
CN116740056A (en) * 2023-08-10 2023-09-12 梁山水泊胶带股份有限公司 Defect detection method for coating layer of whole-core high-pattern conveyer belt
CN116883407A (en) * 2023-09-08 2023-10-13 山东省永星食品饮料有限公司 Bottled water impurity detection method based on artificial intelligence
CN116883415A (en) * 2023-09-08 2023-10-13 东莞市旺佳五金制品有限公司 Thin-wall zinc alloy die casting quality detection method based on image data
CN116912254A (en) * 2023-09-14 2023-10-20 山东博诚电气有限公司 Cable defect identification method based on data enhancement preprocessing
CN116912261A (en) * 2023-09-15 2023-10-20 张家港市神舟机械有限公司 Plastic mold injection molding surface defect detection method
CN116958136A (en) * 2023-09-19 2023-10-27 惠州市金箭精密部件有限公司 Lead screw thread production defect detection method based on image processing
CN116958144A (en) * 2023-09-20 2023-10-27 东莞市南谷第电子有限公司 Rapid positioning method and system for surface defect area of new energy connecting line
CN116993726A (en) * 2023-09-26 2023-11-03 山东克莱蒙特新材料科技有限公司 Mineral casting detection method and system
CN116993722A (en) * 2023-09-26 2023-11-03 山东奥斯登房车有限公司 Visual detection method for defects of trailer body of trailer
CN116993966A (en) * 2023-09-27 2023-11-03 诺伯特智能装备（山东）有限公司 Casting polishing vision intelligent positioning method and system
CN117058130A (en) * 2023-10-10 2023-11-14 威海威信光纤科技有限公司 Visual inspection method for coating quality of optical fiber drawing surface
CN116740065B (en) * 2023-08-14 2023-11-21 山东伟国板业科技有限公司 Quick tracing method and system for defective products of artificial board based on big data
CN117115153A (en) * 2023-10-23 2023-11-24 威海坤科流量仪表股份有限公司 Intelligent printed circuit board quality detection method based on visual assistance
CN117197741A (en) * 2023-09-12 2023-12-08 杭州万禾电力科技有限公司 Switch cabinet operation abnormity monitoring method based on artificial intelligence
CN117218549A (en) * 2023-11-07 2023-12-12 山东力加力钢结构有限公司 State evaluation method and system for highway bridge
CN117237333A (en) * 2023-11-07 2023-12-15 江苏森力威冶金设备有限公司 Stainless steel material surface pickling degree detection method
CN117314912A (en) * 2023-11-29 2023-12-29 山东贞元汽车车轮有限公司 Visual detection method and system for welding defects on surface of welded pipe
CN117314920A (en) * 2023-11-30 2023-12-29 江苏高特高金属科技有限公司 Method for detecting weld flash on surface of X-ray weld joint of metal pipeline
CN117437233A (en) * 2023-12-21 2024-01-23 山东润通齿轮集团有限公司 Gear defect detection method and system based on image processing
CN117455916A (en) * 2023-12-25 2024-01-26 山东太阳耐磨件有限公司 Visual detection method for surface defects of steel plate
CN117455908A (en) * 2023-12-22 2024-01-26 山东济矿鲁能煤电股份有限公司阳城煤矿 Visual detection method and system for belt conveyor deviation
CN117455800A (en) * 2023-12-22 2024-01-26 肥城新查庄地质勘查有限公司 Coal mine drilling positioning method based on computer vision
CN117557566A (en) * 2024-01-12 2024-02-13 深圳博上科技有限公司 Intelligent detection method for quality of heating wire of electronic cigarette based on image characteristics
CN117593303A (en) * 2024-01-18 2024-02-23 浙江锦德光电材料有限公司 Defect detection method and system for quantum dot optical film
CN117611583A (en) * 2024-01-22 2024-02-27 张家港飞腾复合新材料股份有限公司 Artificial intelligence-based aluminum composite panel defect detection method and system
CN117635604A (en) * 2024-01-23 2024-03-01 华航检测认证(青岛)有限公司 Method for detecting welding quality of constructional engineering steel structure
CN117689659A (en) * 2024-02-02 2024-03-12 深圳市未尔科技有限公司 Production quality monitoring method based on flat electronic product
CN117723548A (en) * 2023-12-14 2024-03-19 东莞市毅廷音响科技有限公司 Automobile horn production quality detection method
CN117764992A (en) * 2024-02-22 2024-03-26 山东乔泰管业科技有限公司 Plastic pipe quality detection method based on image processing
CN117764952A (en) * 2023-12-25 2024-03-26 卡尔德线缆(东莞)有限公司 High-speed cable quality detection method for vision robot
CN117788464A (en) * 2024-02-26 2024-03-29 卡松科技股份有限公司 Industrial gear oil impurity visual detection method
CN117876360A (en) * 2024-03-08 2024-04-12 卡松科技股份有限公司 Intelligent detection method for lubricating oil quality based on image processing
CN117893532A (en) * 2024-03-14 2024-04-16 山东神力索具有限公司 Die crack defect detection method for die forging rigging based on image processing
CN117893541A (en) * 2024-03-18 2024-04-16 济南玖通志恒信息技术有限公司 Fruit tree leaf mosaic analysis method based on edge detection
CN117975374A (en) * 2024-03-29 2024-05-03 山东天意机械股份有限公司 Intelligent visual monitoring method for double-skin wall automatic production line
CN117974644A (en) * 2024-03-29 2024-05-03 山东中泰药业有限公司 Visual detection method for quality of capsule medicine
CN118014988A (en) * 2024-03-28 2024-05-10 浙江康鹏半导体有限公司 Intelligent gallium arsenide substrate wafer defect detection method
CN118071753A (en) * 2024-04-24 2024-05-24 森骏卓越精密智造(深圳)有限公司 Visual inspection method for surface defects of injection mold
CN118096736A (en) * 2024-04-23 2024-05-28 山东德源电力科技股份有限公司 Shell surface defect detection method for concentrator terminal production
CN118134924A (en) * 2024-05-07 2024-06-04 浙江富昌机械有限公司 Intelligent detection method for removing waste capsules by hard capsule filling machine based on image processing
CN118135337A (en) * 2024-05-06 2024-06-04 江苏裕荣光电科技有限公司 Cable temperature anomaly identification method based on feature extraction

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115375688B (en) * 2022-10-25 2023-04-28 苏州德斯米尔智能科技有限公司 Automatic detection method for belt conveyor
CN115861291B (en) * 2023-01-06 2023-05-09 曲阜天博汽车电器有限公司 Chip circuit board production defect detection method based on machine vision
CN116091499B (en) * 2023-04-07 2023-06-20 山东中胜涂料有限公司 Abnormal paint production identification system
CN116503397B (en) * 2023-06-26 2023-09-01 山东天通汽车科技股份有限公司 In-vehicle transmission belt defect detection method based on image data
CN116630332B (en) * 2023-07-26 2023-09-26 山东华航高分子材料有限公司 PVC plastic pipe orifice defect detection method based on image processing
CN117611589B (en) * 2024-01-24 2024-04-12 深圳市格瑞邦科技有限公司 Tablet personal computer quality detection method and system

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190228515A1 (en) * 2018-01-22 2019-07-25 Boe Technology Group Co., Ltd. Method and apparatus for detecting defects, and computer readable storage medium
CN114419025A (en) * 2022-01-27 2022-04-29 江苏泰和木业有限公司 Fiberboard quality evaluation method based on image processing
CN114972357A (en) * 2022-08-03 2022-08-30 南通恒立机械设备有限公司 Roller surface defect detection method and system based on image processing
CN115082482A (en) * 2022-08-23 2022-09-20 山东优奭趸泵业科技有限公司 Metal surface defect detection method
CN115375688A (en) * 2022-10-25 2022-11-22 苏州德斯米尔智能科技有限公司 Automatic detection method for belt type conveying equipment

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114998198A (en) * 2022-04-24 2022-09-02 南通夏克塑料包装有限公司 Injection molding surface defect identification method
CN114937039B (en) * 2022-07-21 2022-10-25 阿法龙(山东)科技有限公司 Intelligent detection method for steel pipe defects

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190228515A1 (en) * 2018-01-22 2019-07-25 Boe Technology Group Co., Ltd. Method and apparatus for detecting defects, and computer readable storage medium
CN114419025A (en) * 2022-01-27 2022-04-29 江苏泰和木业有限公司 Fiberboard quality evaluation method based on image processing
CN114972357A (en) * 2022-08-03 2022-08-30 南通恒立机械设备有限公司 Roller surface defect detection method and system based on image processing
CN115082482A (en) * 2022-08-23 2022-09-20 山东优奭趸泵业科技有限公司 Metal surface defect detection method
CN115375688A (en) * 2022-10-25 2022-11-22 苏州德斯米尔智能科技有限公司 Automatic detection method for belt type conveying equipment

Cited By (93)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116630315A (en) * 2023-07-24 2023-08-22 山东东阿亿福缘阿胶制品有限公司 Intelligent beverage packaging defect detection method based on computer vision
CN116630315B (en) * 2023-07-24 2023-09-29 山东东阿亿福缘阿胶制品有限公司 Intelligent beverage packaging defect detection method based on computer vision
CN116630815B (en) * 2023-07-25 2023-09-22 济南玖通志恒信息技术有限公司 Intelligent agricultural pest detection method
CN116630815A (en) * 2023-07-25 2023-08-22 济南玖通志恒信息技术有限公司 Intelligent agricultural pest detection method
CN116630329A (en) * 2023-07-26 2023-08-22 山东山森数控技术有限公司 Online visual detection method for multi-axis multi-channel numerical control system
CN116630329B (en) * 2023-07-26 2023-09-29 山东山森数控技术有限公司 Online visual detection method for multi-axis multi-channel numerical control system
CN116681695B (en) * 2023-07-27 2023-12-01 山东阁林板建材科技有限公司 Quality detection method for anti-deformation template end face
CN116681695A (en) * 2023-07-27 2023-09-01 山东阁林板建材科技有限公司 Quality detection method for anti-deformation template end face
CN116664569A (en) * 2023-07-31 2023-08-29 山东正华建筑科技有限公司 Weld flash defect detection method
CN116664569B (en) * 2023-07-31 2023-10-10 山东正华建筑科技有限公司 Weld flash defect detection method
CN116682107A (en) * 2023-08-03 2023-09-01 山东国宏生物科技有限公司 Soybean visual detection method based on image processing
CN116682107B (en) * 2023-08-03 2023-10-10 山东国宏生物科技有限公司 Soybean visual detection method based on image processing
CN116740056A (en) * 2023-08-10 2023-09-12 梁山水泊胶带股份有限公司 Defect detection method for coating layer of whole-core high-pattern conveyer belt
CN116740056B (en) * 2023-08-10 2023-11-07 梁山水泊胶带股份有限公司 Defect detection method for coating layer of whole-core high-pattern conveyer belt
CN116740057A (en) * 2023-08-11 2023-09-12 深圳市鹏基精密工业有限公司 Cylindrical workpiece burr online detection method and system
CN116721106A (en) * 2023-08-11 2023-09-08 山东明达圣昌铝业集团有限公司 Profile flaw visual detection method based on image processing
CN116740057B (en) * 2023-08-11 2023-12-01 深圳市鹏基精密工业有限公司 Cylindrical workpiece burr online detection method and system
CN116721106B (en) * 2023-08-11 2023-10-20 山东明达圣昌铝业集团有限公司 Profile flaw visual detection method based on image processing
CN116740065B (en) * 2023-08-14 2023-11-21 山东伟国板业科技有限公司 Quick tracing method and system for defective products of artificial board based on big data
CN116735612B (en) * 2023-08-15 2023-11-07 山东精亿机械制造有限公司 Welding defect detection method for precise electronic components
CN116735612A (en) * 2023-08-15 2023-09-12 山东精亿机械制造有限公司 Welding defect detection method for precise electronic components
CN116883407B (en) * 2023-09-08 2023-11-24 山东省永星食品饮料有限公司 Bottled water impurity detection method based on artificial intelligence
CN116883407A (en) * 2023-09-08 2023-10-13 山东省永星食品饮料有限公司 Bottled water impurity detection method based on artificial intelligence
CN116883415A (en) * 2023-09-08 2023-10-13 东莞市旺佳五金制品有限公司 Thin-wall zinc alloy die casting quality detection method based on image data
CN116883415B (en) * 2023-09-08 2024-01-05 东莞市旺佳五金制品有限公司 Thin-wall zinc alloy die casting quality detection method based on image data
CN117197741B (en) * 2023-09-12 2024-02-13 杭州万禾电力科技有限公司 Switch cabinet operation abnormity monitoring method based on artificial intelligence
CN117197741A (en) * 2023-09-12 2023-12-08 杭州万禾电力科技有限公司 Switch cabinet operation abnormity monitoring method based on artificial intelligence
CN116912254B (en) * 2023-09-14 2023-12-08 山东博诚电气有限公司 Cable defect identification method based on data enhancement preprocessing
CN116912254A (en) * 2023-09-14 2023-10-20 山东博诚电气有限公司 Cable defect identification method based on data enhancement preprocessing
CN116912261A (en) * 2023-09-15 2023-10-20 张家港市神舟机械有限公司 Plastic mold injection molding surface defect detection method
CN116912261B (en) * 2023-09-15 2023-12-08 张家港市神舟机械有限公司 Plastic mold injection molding surface defect detection method
CN116958136A (en) * 2023-09-19 2023-10-27 惠州市金箭精密部件有限公司 Lead screw thread production defect detection method based on image processing
CN116958136B (en) * 2023-09-19 2024-01-30 惠州市金箭精密部件有限公司 Lead screw thread production defect detection method based on image processing
CN116958144B (en) * 2023-09-20 2024-01-12 东莞市南谷第电子有限公司 Rapid positioning method and system for surface defect area of new energy connecting line
CN116958144A (en) * 2023-09-20 2023-10-27 东莞市南谷第电子有限公司 Rapid positioning method and system for surface defect area of new energy connecting line
CN116993722B (en) * 2023-09-26 2023-12-08 山东奥斯登房车有限公司 Visual detection method for defects of trailer body of trailer
CN116993726A (en) * 2023-09-26 2023-11-03 山东克莱蒙特新材料科技有限公司 Mineral casting detection method and system
CN116993722A (en) * 2023-09-26 2023-11-03 山东奥斯登房车有限公司 Visual detection method for defects of trailer body of trailer
CN116993726B (en) * 2023-09-26 2023-12-19 山东克莱蒙特新材料科技有限公司 Mineral casting detection method and system
CN116993966A (en) * 2023-09-27 2023-11-03 诺伯特智能装备（山东）有限公司 Casting polishing vision intelligent positioning method and system
CN116993966B (en) * 2023-09-27 2023-12-12 诺伯特智能装备（山东）有限公司 Casting polishing vision intelligent positioning method and system
CN117058130B (en) * 2023-10-10 2024-01-09 威海威信光纤科技有限公司 Visual inspection method for coating quality of optical fiber drawing surface
CN117058130A (en) * 2023-10-10 2023-11-14 威海威信光纤科技有限公司 Visual inspection method for coating quality of optical fiber drawing surface
CN117115153A (en) * 2023-10-23 2023-11-24 威海坤科流量仪表股份有限公司 Intelligent printed circuit board quality detection method based on visual assistance
CN117115153B (en) * 2023-10-23 2024-02-02 威海坤科流量仪表股份有限公司 Intelligent printed circuit board quality detection method based on visual assistance
CN117237333A (en) * 2023-11-07 2023-12-15 江苏森力威冶金设备有限公司 Stainless steel material surface pickling degree detection method
CN117218549B (en) * 2023-11-07 2024-03-05 山东力加力钢结构有限公司 State evaluation method and system for highway bridge
CN117237333B (en) * 2023-11-07 2024-02-20 江苏森力威冶金设备有限公司 Stainless steel material surface pickling degree detection method
CN117218549A (en) * 2023-11-07 2023-12-12 山东力加力钢结构有限公司 State evaluation method and system for highway bridge
CN117314912A (en) * 2023-11-29 2023-12-29 山东贞元汽车车轮有限公司 Visual detection method and system for welding defects on surface of welded pipe
CN117314912B (en) * 2023-11-29 2024-01-30 山东贞元汽车车轮有限公司 Visual detection method and system for welding defects on surface of welded pipe
CN117314920A (en) * 2023-11-30 2023-12-29 江苏高特高金属科技有限公司 Method for detecting weld flash on surface of X-ray weld joint of metal pipeline
CN117314920B (en) * 2023-11-30 2024-02-02 江苏高特高金属科技有限公司 Method for detecting weld flash on surface of X-ray weld joint of metal pipeline
CN117723548A (en) * 2023-12-14 2024-03-19 东莞市毅廷音响科技有限公司 Automobile horn production quality detection method
CN117437233A (en) * 2023-12-21 2024-01-23 山东润通齿轮集团有限公司 Gear defect detection method and system based on image processing
CN117437233B (en) * 2023-12-21 2024-03-26 山东润通齿轮集团有限公司 Gear defect detection method and system based on image processing
CN117455800A (en) * 2023-12-22 2024-01-26 肥城新查庄地质勘查有限公司 Coal mine drilling positioning method based on computer vision
CN117455908B (en) * 2023-12-22 2024-04-09 山东济矿鲁能煤电股份有限公司阳城煤矿 Visual detection method and system for belt conveyor deviation
CN117455908A (en) * 2023-12-22 2024-01-26 山东济矿鲁能煤电股份有限公司阳城煤矿 Visual detection method and system for belt conveyor deviation
CN117455800B (en) * 2023-12-22 2024-03-15 肥城新查庄地质勘查有限公司 Coal mine drilling positioning method based on computer vision
CN117455916A (en) * 2023-12-25 2024-01-26 山东太阳耐磨件有限公司 Visual detection method for surface defects of steel plate
CN117764952B (en) * 2023-12-25 2024-05-14 卡尔德线缆(东莞)有限公司 High-speed cable quality detection method for vision robot
CN117764952A (en) * 2023-12-25 2024-03-26 卡尔德线缆(东莞)有限公司 High-speed cable quality detection method for vision robot
CN117455916B (en) * 2023-12-25 2024-03-15 山东太阳耐磨件有限公司 Visual detection method for surface defects of steel plate
CN117557566B (en) * 2024-01-12 2024-04-12 深圳博上科技有限公司 Intelligent detection method for quality of heating wire of electronic cigarette based on image characteristics
CN117557566A (en) * 2024-01-12 2024-02-13 深圳博上科技有限公司 Intelligent detection method for quality of heating wire of electronic cigarette based on image characteristics
CN117593303A (en) * 2024-01-18 2024-02-23 浙江锦德光电材料有限公司 Defect detection method and system for quantum dot optical film
CN117593303B (en) * 2024-01-18 2024-04-09 浙江锦德光电材料有限公司 Defect detection method and system for quantum dot optical film
CN117611583A (en) * 2024-01-22 2024-02-27 张家港飞腾复合新材料股份有限公司 Artificial intelligence-based aluminum composite panel defect detection method and system
CN117611583B (en) * 2024-01-22 2024-04-19 张家港飞腾复合新材料股份有限公司 Artificial intelligence-based aluminum composite panel defect detection method and system
CN117635604A (en) * 2024-01-23 2024-03-01 华航检测认证(青岛)有限公司 Method for detecting welding quality of constructional engineering steel structure
CN117635604B (en) * 2024-01-23 2024-04-05 华航检测认证(青岛)有限公司 Method for detecting welding quality of constructional engineering steel structure
CN117689659A (en) * 2024-02-02 2024-03-12 深圳市未尔科技有限公司 Production quality monitoring method based on flat electronic product
CN117689659B (en) * 2024-02-02 2024-04-19 深圳市未尔科技有限公司 Production quality monitoring method based on flat electronic product
CN117764992A (en) * 2024-02-22 2024-03-26 山东乔泰管业科技有限公司 Plastic pipe quality detection method based on image processing
CN117764992B (en) * 2024-02-22 2024-04-30 山东乔泰管业科技有限公司 Plastic pipe quality detection method based on image processing
CN117788464A (en) * 2024-02-26 2024-03-29 卡松科技股份有限公司 Industrial gear oil impurity visual detection method
CN117788464B (en) * 2024-02-26 2024-04-30 卡松科技股份有限公司 Industrial gear oil impurity visual detection method
CN117876360A (en) * 2024-03-08 2024-04-12 卡松科技股份有限公司 Intelligent detection method for lubricating oil quality based on image processing
CN117893532A (en) * 2024-03-14 2024-04-16 山东神力索具有限公司 Die crack defect detection method for die forging rigging based on image processing
CN117893532B (en) * 2024-03-14 2024-05-24 山东神力索具有限公司 Die crack defect detection method for die forging rigging based on image processing
CN117893541A (en) * 2024-03-18 2024-04-16 济南玖通志恒信息技术有限公司 Fruit tree leaf mosaic analysis method based on edge detection
CN117893541B (en) * 2024-03-18 2024-05-28 济南玖通志恒信息技术有限公司 Fruit tree leaf mosaic analysis method based on edge detection
CN118014988B (en) * 2024-03-28 2024-06-07 浙江康鹏半导体有限公司 Intelligent gallium arsenide substrate wafer defect detection method
CN118014988A (en) * 2024-03-28 2024-05-10 浙江康鹏半导体有限公司 Intelligent gallium arsenide substrate wafer defect detection method
CN117975374A (en) * 2024-03-29 2024-05-03 山东天意机械股份有限公司 Intelligent visual monitoring method for double-skin wall automatic production line
CN117974644A (en) * 2024-03-29 2024-05-03 山东中泰药业有限公司 Visual detection method for quality of capsule medicine
CN117974644B (en) * 2024-03-29 2024-06-11 山东中泰药业有限公司 Visual detection method for quality of capsule medicine
CN118096736A (en) * 2024-04-23 2024-05-28 山东德源电力科技股份有限公司 Shell surface defect detection method for concentrator terminal production
CN118071753A (en) * 2024-04-24 2024-05-24 森骏卓越精密智造(深圳)有限公司 Visual inspection method for surface defects of injection mold
CN118135337A (en) * 2024-05-06 2024-06-04 江苏裕荣光电科技有限公司 Cable temperature anomaly identification method based on feature extraction
CN118135337B (en) * 2024-05-06 2024-07-05 江苏裕荣光电科技有限公司 Cable temperature anomaly identification method based on feature extraction
CN118134924A (en) * 2024-05-07 2024-06-04 浙江富昌机械有限公司 Intelligent detection method for removing waste capsules by hard capsule filling machine based on image processing

Also Published As

Publication number Publication date
LU504264B1 (en) 2023-07-31
CN115375688A (en) 2022-11-22
CN115375688B (en) 2023-04-28

Similar Documents

Publication Publication Date Title
WO2023134789A1 (en) Automatic inspection method for belt-type conveying device
CN115239735B (en) Communication cabinet surface defect detection method based on computer vision
CN116703907B (en) Machine vision-based method for detecting surface defects of automobile castings
CN115082418B (en) Precise identification method for automobile parts
CN114972356B (en) Plastic product surface defect detection and identification method and system
CN116503397B (en) In-vehicle transmission belt defect detection method based on image data
CN114693610A (en) Welding seam surface defect detection method, equipment and medium based on machine vision
CN115953398B (en) Defect identification method for strip steel surface
CN115063423A (en) Mechanical casting cold and hot crack self-adaptive identification method based on computer vision
CN114820612B (en) Roller surface defect detection method and system based on machine vision
CN116091504A (en) Connecting pipe connector quality detection method based on image processing
CN116563279B (en) Measuring switch detection method based on computer vision
CN101995412A (en) Robust glass scratch defect detection method and device thereof
CN115359053A (en) Intelligent detection method and system for defects of metal plate
CN115511842A (en) Cable insulation skin damage detection method based on machine vision
CN115797361B (en) Aluminum template surface defect detection method
CN115631146A (en) Image-based pantograph carbon slide strip defect detection method
CN114612403A (en) Intelligent detection method and system for breakage defect of feeding belt
CN117333727A (en) Conveyer belt wearing and tearing area discernment analysis system based on visual image processing technique
CN115159027B (en) Belt tearing monitoring method
CN107024416A (en) With reference to similitude and the quasi-circular particle mean size detection method of discontinuity
CN115018792A (en) Deep-drawing part wrinkling detection method based on computer vision technology
Yao et al. Automated detection and identification of area-based distress in concrete pavements
CN115049641A (en) Electric data processing method and system for anomaly detection of mechanical parts
CN114972353A (en) Quality evaluation method and system for metal high-pressure special pipe fitting

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 23740129

Country of ref document: EP

Kind code of ref document: A1