The Research On Portrait Background Segmentation Algorithm

Portrait background segmentation applies the color image segmentation algorithms to the portrait images, in order to segment the portrait foreground from the background, and change the color of the background at the meantime. It can be widely used for the systems of the certificate management, photo process in the department of personnel, police and custom, etc.This paper analyses and compares the research actuality of the color image segmentation algorithms, and then applies an improved Fuzzy C-Means Clustering (FFCM) to the portrait background segmentation. First, It uses the hierarchical clustering, which could reduce the heavy computation load when clustering a large number of data points, to partition the image data into a certain number of subsets with similar color. Then, these centers of the subsets and the number of points in the neighborhood of centers are used in FCM. Besides, this paper proposes a method of manual object selection, which could quickly revert the original portrait object which having been fuzzy segmented, and change...