Matching of Bag’s Images Based on Straight Line Feature Control

Abstract

Proposed matching method of bag’s images based on straight line features control in this paper. At first, it established a mathematical model of straight line features control theory; Secondly, it collected bags images by use of machine vision systems. It applied the contrast, image binarization, the first order derivative edge detection, dilation and erosion, image thinning, noise remove and other image processing technology and got the edge of the target object image. Next it used the Harris corner detection algorithm to get corners of edge image which is needed by straight linear control theory and achieve the features matching of two bag’s edge images. Finally, it processed vertical, rotating, tilt, moving and light changing’s bags’ images when putting bags stacked and side by side, the results show that the proposed matching method has strong robustness.