Weld surface defect feature extraction method based on image gray scale B sweeping curve

Abstract

The invention discloses a weld surface defect feature extraction method based on an image gray scale B sweeping curve, and the method comprises the steps: setting photographing parameters of a miniature CCD camera according to an image collection standard; converting a collected true color image into a gray scale map, and carrying out the median filtering of the gray scale map; carrying out the preliminary extraction of the edge lines between a weld region and a background and between a non-welding region and the background in the gray scale map through employing a Canny operator edge line extraction algorithm based on a threshold value; judging the welding surface defects according to the number and shapes of edge lines, such as non-centered welding and weld bead distortion; reconstructing a continuous edge line of the weld region through Hough transformation, and achieving the accurate positioning of the weld region; drawing a section gray scale B sweeping curve perpendicular to the weld edge, wherein the difference of gray scale values of the B sweeping curve at two ends of the weld changes remarkably when there is misalignment on the weld surface, so as to judge the misalignment phenomenon of the weld surface. The method achieves the accurate positioning of the weld edge and the accurate recognition of defects of non-centered welding, weld bead distortion and misalignment.

Description

technical field [0001] The invention relates to a detection method for weld surface defects, in particular to a weld defect recognition method based on image processing technology. The method is suitable for the detection and identification of weld surface defects in automatic welding processes, and belongs to the field of non-destructive testing. Background technique [0002] Welding technology is widely used in industrial production. At present, large workpieces such as boilers and pipelines are mostly completed by automatic welding technology. In the production line automatic welding process, it is inevitable that there will be defects such as misalignment, partial welding and broken welding that affect product quality. The existence of weld surface defects will greatly reduce the safety of welded products during the service period, ranging from product failure to leakage, to serious causes of brittle fracture of the product, causing serious casualties. Therefore, the d...

AI Technical Summary

Problems solved by technology

[0005] At present, the technology of welding seam inspection based on X-ray is mostly for the detection of internal defects of weld seam welding, and there is little research on the image detection and recognition technology of weld seam surface defects

Images