基于聚类算法的人脸衰老图像重构

Aging Face Image Synthesis Based on Clustering Algorithm

本文基于人脸图形图像数据库,提出了基于聚类算法的人脸衰老图像合成方法。采用改进主动形状模型(ASM)的人脸特征点匹配算法,对面部轮廓特征点曲率进行分类,获得同一簇内的人脸面部轮廓较高相似度。结合小波变换技术对人脸图形图像光泽的变化、皱纹的增加等衰老的处理,实现人脸图形图像的衰老变化。

A synthesis method of facial aging image based on clustering algorithm is proposed by using the face image database in this paper. A high similarity of face contour in the same cluster is obtained in classifying process of facial contour feature by using face feature point matching algorithm of improved active shape model（ ASM）. Combined with wavelet transform technology,he aging changes of the facial image are realized by handling face image of luster,wrinkles and aging.