Features extraction for structured light image of welding seam with arc and splash disturbance

A method of image process and features extraction for structured light image of welding seam with arc and splash disturbance is proposed. The seam area is detected by search with large step. The adaptive thresholds of image enhancement are determined in the frequency domain of the gray image. Then, the target image is pre-processed using image enhancement and binarization. After thinning the seam with its both edges, main characteristic line is obtained using Hotelling transform and Hough transform. Finally, the feature points in the seam are found according to its second derivative. Experimental results show its effectiveness, good performance in real time and adaptability to different seams.