Nondestructive detection method for weld defects based on computer vision

Abstract

The invention relates to a nondestructive detection method for weld defects based on computer vision, which comprises the following steps of carrying out semantic segmentation on a weld x-ray negative film to obtain a weld edge, and carrying out connected domain analysis on an inner side area of the weld edge to obtain a connected domain on the inner side of the weld; segmenting the edge of each connected domain to obtain an outer edge close to one side of the base material and an inner edge close to one side of the center of the welding seam; respectively obtaining the identification degree, the blackness and the straightness of the inner edge and the outer edge; obtaining the definition of the inner edge and the outer edge by using the identification degree and the blackness; obtaining a non-fusion rate of the connected domain according to a definition difference value and a straightness difference value between the inner edge and the outer edge; and setting a threshold value, comparing the threshold value with the non-fusion rate corresponding to the connected domain, and marking the connected domain according to a comparison result of the threshold value and the non-fusion rate corresponding to the connected domain. According to the technical means provided by the invention, the non-fusion defect of the welding seam can be accurately identified, and the easy-to-confuse type identification of the welding seam defect is more accurate.

Description

technical field [0001] The invention relates to the field of computer vision, in particular to a computer vision-based nondestructive detection method for weld defects. Background technique [0002] In the welding process, various welding defects will be generated due to various factors such as improper operation or unqualified welding materials. In order to detect these defects, the prior art uses X-rays to irradiate the weld seam from top to bottom According to the absorption of x-rays by the weld metal to obtain the x-ray base plate of the weld, non-destructive testing can be realized, that is, the internal defects of the weld can be detected without destroying the weld. When a weld is flawed, features such as shadows of different shapes are formed on it. Based on these features, the operator can identify defects in the weld. [0003] The defects on the existing x-ray film of the welding seam are often directly identified manually, which is inefficient. Other neural ne...

AI Technical Summary

Problems solved by technology

These two types of defects are similar in shape, and if they are confused, they will cause misjudgment of the degree of damage to the weld and cause danger.

Images