摘要
为了提高人脸的识别率和识别速度及其识别的鲁棒性,提出了基于拓展稀疏表示模型和LC-KSVD(Label Consist K-SVD)的人脸识别算法。针对字典学习中只包含表示能力没有包含类别信息的问题,在原始的稀疏表示模型中添加了残差向量作为系数修正向量,使得拓展稀疏表示模型具有更强的鲁棒性;在字典学习中添加稀疏编码和分类器参数约束项,通过字典学习同时更新稀疏编码和分类器参数,使字典中包含很好的表示能力和判别分类能力。实验结果表明,基于拓展稀疏表示模型和LC-KSVD的人脸识别具有高识别率和低识别速度,并且有很好的鲁棒性。
To improve the face recognition rate, speed and robustness, this paper proposes a face recognition algorithmbased on extended sparse representation model and LC-KSVD(Label Consist K-SVD). For solving the problem that dictionarylearning only contains representation ability but no class information, the algorithm adds residual vector as coefficientamending vector into original sparse representation model, making the extended model have stronger robustness.The algorithm also adds sparse coding and classifier parameter constraints into the process of dictionary learning andupdates sparse coding and classifier parameters in the process, making the dictionary possess good representation and discriminationability. The experimental results show that the algorithm has high recognition rate, low recognition speed andgood robustness.
出处
《计算机工程与应用》
CSCD
北大核心
2016年第13期206-211,共6页
Computer Engineering and Applications
基金
国家自然科学基金(No.61202439)
湖南省教育厅优秀青年项目(No.12B003)
湖南省交通运输厅科技进步与创新项目(No.201334)
关键词
稀疏表示
字典学习
人脸识别
LC-KSVD算法
sparse representation
dictionary learning
face representation
LC-KSVD(Label Consist K-SVD)