Seam Detection Welding System Based on Image Segmentation with Multi-feature Fusion
Welding seam inspection systems have the issues of susceptibility to lighting conditions,reliance on manual operation,and high risk of collision accidents in industrial production,in response to these problems,an automated welding seam detection sys-tem based on multi-feature fusion segmentation algorithm is developed.The system employs a laser depth camera to capture depth im-ages of weld seams.It combines the multi-feature fusion with the skew moment envelope fitting in the image segmentation algorithm to overcome the challenges posed by uneven surfaces.The accurate acquisition of weld seam coordinates and width is ensured through the calibration algorithm.Additionally,a collision avoidance detection algorithm based on oriented bounding boxes(OBB)is proposed to prevent collisions during welding process.The system is built by using a robotic arm.Experimental results demonstrate that the system has an effectiveness in detecting weld seam width and positioning,with an average error of 0.302 9 mm and 0.339 3 mm,re-spectively,meeting industrial application standards.This system optimizes the welding path and collision detection,significantly en-hancing welding quality and production efficiency.
NETL
NSTL
万方数据
