Railway vehicle running gear welding seam detection method and system

Abstract

The invention discloses a railway vehicle running gear welding seam detection method. The method comprises the following steps: acquiring a detection image; performing down-sampling based on the detection image and generating a multi-scale feature map; fusing the plurality of feature maps and generating a plurality of fused feature maps; extracting defect category information and defect position information based on the fused feature map; and generating a detection report based on the defect category information and the defect position information. The multi-scale feature maps contained in the detection image are extracted, and feature fusion is carried out on the feature maps of different scales by utilizing the characteristic that the semantic information contained in the feature maps of different scales is different in strength, so that on one hand, complementary utilization of the semantic information contained in the feature maps of different scales is realized, and on the other hand, the influence that unreliable data are introduced or important target features are lost in the detection image preprocessing process due to the fact that the size of the detection image is uncertain is avoided, and the image feature recognition capacity for detecting the tiny defects of the welding seam is improved.

Description

technical field [0001] The invention relates to the technical field of rail vehicle detection, in particular to a method and system for detecting weld seams of running parts of rail vehicles. Background technique [0002] As a key component of rail transit vehicles, the running part plays an important role in train operation. However, due to the incomplete welding seam welding process of the running part, for example, the welding seam welding process of the bogie, wheel set and side frame of the running part is still incomplete, which will cause the vehicle to be affected by the unevenness of the line or the wheel rail during driving. It is easy to cause the fatigue problem of the running part. Taking the bogie structure as an example, if the fatigue defects reflected in the welds of the frame are not detected timely and effectively for a long time, it will induce structural safety hazards of the frame, and the serious consequences will directly affect the driving safety of ...

AI Technical Summary

Problems solved by technology

[0005] In view of this, the present invention provides a method and system for detecting the weld seam of the running part of a rail vehicle. By improving the eddy current detection method of the running part weld and the processing method of the detection data, the problem of the traditional welding seam of the running part of the rail vehicle is solved. The problem of low accuracy in the detection method

Images