Abstract:Image sensor has been one of the key technologies in intellectualized robotics welding. Edge detection plays an important role when the vision technology is applied in intellectualized welding robotics technologies. There are all kinds of noises in welding environment. The algorithms in common use cannot be applied to the recognition of welding environment directly. The edge of images can be fell into four types. The weld images are classified by the characteristic of welding environment in this paper. This paper analyzes some algorithms of edge detection according to the character of welding image, some relative advantages and disadvantages are pointed out when these algorithms are used in this field, and some suggestions are given. The feature extraction of weld seam and weld pool are two typical problems in the realization of intellectualized welding. Their edge features are extracted and the results show the applicability of different edge detectors. The tradeoff between precision and calculated time is also considered for different application.

Edge detection and its application to recognition of arc weld image