Abstract: In this paper, a new closed contour extraction method based on Gestalt rules is proposed to reduce the impact on the background of the edge. Firstly, the integration significant constraint Canny edge detection method, which integrates significant constraints in the framework of Canny edge detection, is given to reduce the background edge in edge extraction process. Then, edge information into the edge fitting algorithms about the fold approximation is proposed to reduce the effects of noise to the edge by Canny edge detection method. Moreover, the edge of the fitting is measured by Closed Rules in Gestalt rules, which can obtain the relationship between the edges. Finally, the new objective function is proposed by analyzing the relationship between the edges and the significance of the target area, the objective function is solved by the minimum weight perfect matching algorithms. Experimental results show that our algorithm improves the effectiveness and accuracy of the contour extraction.