Most conventional derivative-based corner detectors have shortcomings such as missing junctions, poor localization, sensitivity to noise and high computational cost. This paper presents a new, simple and effective low level processing method to detect corners. In this corner detection algorithm, two oriented cross operators called COP (Crosses as Oriented Pair) are used, which provide useful information to extract low-level features due to its characteristics, preference for edge with different direction and simple direction determination. Fast, accurate and noise-robust corner detection is accomplished with the COP.
An image processing method using the COP operator is employed to detect features of the weld seam in arc welding. Hough transform is applied to the feature points and used to recognize the seam location, tack weld, joint gap and multiple seams. The stereo image captured by the biprism vision sensor provides the three-dimensional information. The proposed image processing method appears to be adequate to automation of the arc welding.