Computer Engineering and Applications ›› 2013, Vol. 49 ›› Issue (6): 167-169.

Previous Articles     Next Articles

Facial feature location based on multipopulation genetic algorithm

ZHANG Yong, ZHANG Yiyun, WANG Fang   

  1. School of Computer and Communication, Lanzhou University of Technology, Lanzhou 730050, China
  • Online:2013-03-15 Published:2013-03-14

基于多种群遗传算法的人脸特征定位

张  永,张译匀,王  芳   

  1. 兰州理工大学 计算机与通信学院,兰州 730050

Abstract: Active Shape Model(ASM) which is used to locate feature is a statistics shape model. Based on the original active shape model, a new Multipopulation Genetic Algorithm(MPGA) is applied to searching for the best representation of face images. The fitness function is determined based on the characteristics of the main facial features. Experimental results show that the improved ASM improves the accuracy of facial features location.

Key words: facial feature location, active shape model, multipopulation genetic algorithm

摘要: 活动形状模型(Active Shape Model,ASM)是一种用于特征定位的统计形状模型。在原活动形状模型的基础上,提出一种新的多种群遗传算法(Multipopulation Genetic Algorithm,MPGA)去搜索人脸图片的最好表示。并且根据面部各主要特征的特点确定适应度函数。实验结果表明,改进的ASM对于人脸特征定位有较好的效果。

关键词: 人脸特征定位, 活动形状模型, 多种群遗传算法