A genetic algorithm on welding seam image segmentation

Abstract

Threshold segmentation is widely used in image segmentation because of its simplicity and efficiency. Otsu's method is one of the superior threshold selection methods. When Otsu's method is used for welding seam image segmentation, because the gray level distribution of the welding seam image often appears asymmetry, separating the seam region from the background of image is difficult. Based on welding seam image analysis and Otsu's method, an adaptive threshold selection method using a genetic algorithm is proposed. The experiment results show that the new method is strongly adaptive and efficient for welding seam image segmentation.