An automatic measurement, characterization and classification method and system for steel cracking defects
A technology of automatic measurement and classification methods, applied in the direction of optical testing flaws/defects, etc., can solve the problems of inability to measure, characterize, low efficiency, low precision, etc.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0091] Use the image acquisition module to obtain the original image of the steel defect to be tested. The hardware of the image acquisition module is as follows: figure 2 Shown: steel sample 1, camera equipment 2, computer 3 and printer 4; the computer 4 is inserted with an image acquisition card; the camera equipment 2 is a camera or a professional camera. The specific steps of image acquisition are to use the camera / professional camera to adjust the image to a suitable focal length, take a picture when the image is clearest and store it in the image acquisition card to obtain the original image, which can be used for subsequent image processing.
[0092] Such as image 3 The original image shown is the defect caused by massive inclusions in the hot-rolled plate. First, the original image is denoised by linear average low-pass filtering. The principle is mainly to replace the original gray threshold with the average gray threshold of each threshold element in the threshold...
Embodiment 2
[0129] Such as Figure 9 The original image shown is a star-shaped cracking defect on the surface of the steel plate. The defect shape is complex and manifested as a cross-line feature. It is almost impossible to measure and analyze it using the traditional manual mode, let alone obtain accurate measurement results. . The process of using the present invention to process it is as follows: firstly, the image acquisition module is used to obtain the original image of the steel defect to be tested, and the collected original image is subjected to linear average low-pass filter denoising, and the image after low-pass filter denoising is as follows: Figure 10 Shown; then the filtered image is subjected to local adaptive threshold segmentation, the image after binary segmentation is as follows Figure 11 As shown; set the scale and calibrate the area of each smallest sub-section of the defect, extract the minimum circumscribing rectangle length L of the smallest sub-segment defe...
Embodiment 3
[0132] Such as Figure 15 The original image shown is the defect image of steel plate cracking defects accompanied by sulfide inclusions, and the defects are characterized by the interweaving of linear defects and massive defects. The process of processing it with the present invention is as follows: firstly, the image acquisition module is used to obtain the original image of the steel defect to be tested, and the original image is subjected to linear average low-pass filter denoising, and the image after low-pass filter denoising is as follows: Figure 16 Shown; then the filtered image is subjected to local adaptive threshold segmentation, the image after binary segmentation is as follows Figure 17 As shown; set the scale and calibrate the area of each smallest sub-section of the defect, extract the minimum circumscribing rectangle length L of the smallest sub-segment defect, the minimum circumscribing rectangle width B of the smallest sub-segment defect, and the linear s...
PUM
| Property | Measurement | Unit |
|---|---|---|
| width | aaaaa | aaaaa |
| width | aaaaa | aaaaa |
| width | aaaaa | aaaaa |
Abstract
Description
Claims
Application Information
Login to View More