Study of Welding Seam Recognizition Based on Vision Method
WANG Wei-jie, ZHOU Sheng-yuan, ZHAO Xue-zeng (School of Mechatronics Engineering, Harbin Institute of Technology, Harbin 150001, China)
This paper applies the vision method to carry out the central position and the gap of the welding seam. Firstly, the author designed an auto-focusing method with the Roberts gradient function. The Roberts gradient function has high sensitivity and focusing scale, good stability, high calculating speed. A simple and practicality demarcate method which is based uprightness measure scheme is introduced. It can satisfy the precision of the system. Finally, this paper put forward a series of processing steps for welding seam images which include median filtering, creating binary image by adaptive threshold and acnode filtering are focused. The experiment results show that the measuring system can satisfy the real time request and effectively pick-up the central position and the gap of the welding seam.
【CateGory Index】： TG434