A Location Method for Ceramic Tile Based on the Combination Features

Locating and sorting of ceramic tile is performed by human observations in most factories. To improve the efficiency of ceramic tile location, an image processing method is proposed in this paper. At first, the Gauss pyramid model is constructed to reduce the search space, the segmentation algorithm and 8 neighborhood chain code are adopted to obtain the contour of ceramic tile, and then combination features vector is generated from contour primitive characteristics and color space separately. Finally, the optimized parameters from feature distribution and a discriminant function are utilized to achieve target recognition. The experiment is conducted on an assembly line with two SCARA robots, and the results show that the proposed method is effective and meets the real-time requirements in complex environment.