The New Method for Segmentation of Moving Object

摘要

Segmentation is a useful technique in image analysis. Traditional image processing techniques based on grey scale images have not provided good results in road traffic applications. To extract moving vehicles from a busy background is still difficult. In this paper, we present a new approach to segment the moving object. This technique combines motion segmentation and region growing. Through the motion segmentation, the approximate position of the object can be determined, and most of the background can be removed. By applying the region growing on the position of motion mask, the object can be extracted and the exact contour of the object can be achieved. The experiment results show that this method is quite promising.