Realization of Real-time Edge Detection of Weld Pool for YAG laser Welding in TWB

Due to the disturbances of spatters, dusts and strong arc light, it is difficult to detect the weld pool edge in laser welding processes. An image sensing system for the ND:YAG laser welding process is introduced in detail. Image processing and pattern recognition in the system are first used to obtain information from the laser welding process for tailed weld blanks. A new way is employed to preprocess the image of the laser welding in order to detect effectively the edge of weld pool in tailed weld blanks. The image of the weld pool was processed using a series of methods: image truncation, bi-level thresholding, median filter and edge detection. The experimental results show that better performance to extract the edge of the weld pool can be obtained by using the new way. The proposed edge detection approach can reach not only perfect edge detection result but also good robustness to noise in TWB. Experiments also show that using the welding monitor system to control the ND:YAG laser welding quality for stainless steel is an effective method.