Image processing-based detection method for missing bolts in span beam assembly
An image processing and loss detection technology, applied in the field of image processing, can solve problems such as missing cross beam assembly bolts, and achieve the effects of improving operating efficiency, reducing labor costs, and improving detection efficiency.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
specific Embodiment approach 1
[0040] Specific implementation mode one: the following combination Figure 1 to Figure 3 Describe the present embodiment, the image processing-based method for detecting the loss of assembly bolts of railway freight car cross beams described in this embodiment, the method includes the following steps:
[0041] Step 1. Use the OTSU algorithm to perform threshold segmentation on the spanning beam image, and preliminarily determine the fault identification area;
[0042] Step 2, use the improved canny operator to traverse all the contour points in the fault identification area, determine the boundary points across the beam, and then obtain the point set A across the centerline of the beam in the y direction;
[0043] Step 3. Taking the point with the smallest y value in the point set A as the reference, offset in four directions up, down, left, and right and take a screenshot as the fault identification image;
[0044] Traverse the point set A to obtain the point with the smalle...
specific Embodiment approach 2
[0052] Specific implementation mode two: the following combination figure 2 with image 3 This embodiment is described. This embodiment will further explain Embodiment 1. In step 1, the OTSU algorithm is used to perform threshold segmentation on the spanning beam image, and the process of initially determining the fault identification area is as follows:
[0053] The OTSU algorithm is used to perform threshold segmentation on the cross-beam image to obtain the black shadow area under the bogie hole;
[0054] Extract the contour of the shaded area and obtain the maximum point in the y direction of the contour;
[0055] Based on this point, shift left along the x-axis by two span beam widths, shift right along the x-axis by one span beam width, and shift up by one hole height along the y-axis, and intercept the sub-image of this area as the fault identification area .
[0056] Because the camera is shooting at an elevation angle, a black shadow area is formed in the lower part...
specific Embodiment approach 3
[0058] Specific implementation mode three: the following combination figure 2 with image 3 Describe this implementation mode, this implementation mode will further explain implementation mode 1 or 2, the process of obtaining point set A in step 2 is:
[0059] Use the improved canny operator to traverse all the contour points in the fault identification area to obtain the position of all contour points in the fault identification area and the corresponding gradient size and direction;
[0060] Traversing the contour points, when the gradient of the contour point is greater than the threshold and the angle between the gradient direction and the horizontal direction is less than the threshold, record the contour point as (x1, y1); continue to traverse to find the contour point (x2, y2) that meets the condition , the conditions for the contour point (x2, y2) to be satisfied are: the y value of the point (x1, y1) is the same, the difference between the x value of the point (x1, ...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More