摘要
人脸表情特征选择是人脸表情识别研究领域关注的一个热点。基于量子遗传算法与邻域粗糙集理论,文章提出一种新的人脸表情特征选择方法(Feature Selection based on Neighborhood Rough Set Theoryand Quantum Genetic Algorithm,简称FSNRSTQGA),以邻域粗糙集理论为基础,定义了最优特征集的适应度函数来评价表情特征子集的选择效果;并结合量子遗传算法进化策略,提出了一种表情特征选择方法。Cohn-Kanade表情数据集上的仿真实验结果表明了该方法的有效性。
Facial expression feature selection is one of the hot issues in the field of facial expression rec- ognition. A novel facial expression feature selection method named feature selection based on neigh- borhood rough set theory and quantum genetic algorithm （FSNRSTQGA） is proposed. First, an eval- uation criterion of the optimization expression feature subset is established based on neighborhood rough set theory and used as the fitness function. Then, by quantum genetic algorithm evolutionary strategy, an approach of facial expression feature selection is proposed. The results of the simulation on Cohn-Kanade expression dataset illustrate that the FSNRSTQGA method is effective.
出处
《合肥工业大学学报（自然科学版）》
CAS
CSCD
北大核心
2013年第1期39-42,128,共5页
Journal of Hefei University of Technology：Natural Science
基金
国家自然科学基金资助项目(61202191
71202165)
关键词
邻域粗糙集
特征选择
量子遗传算法
人脸表情识别
neighborhood rough set
feature selection
quantum genetic algorithm
facial expressionrecognition
