摘要
提取人脸图像光照不变量是提高不完备训练样本人脸识别光照鲁棒性的一个有效途径。以往算法分别从不同角度提取人脸图像的高频特征作为光照不变量不能提取完整的人脸本征,具有一定的局限性。从特征级和决策级融合的角度提出了一种基于多特征融合的复杂光照人脸识别方法。所提算法能发挥不同光照不变量的自身优势,明显提高复杂光照人脸识别的光照鲁棒性。Yale B^+和非控光照人脸库的实验结果表明所提算法的有效性。
Extracting illumination invariants from face image is an effective way to improve the robust-illumination performanceof incomplete training sample face recognition.The previous algorithms extracting the high frequency characteristicsof face image from different angles as illumination invariants can’t get integral essential feature of face and havesome limitations.In this paper,from the perspective of feature level and decision level fusion,a complex illumination facerecognition method based on multi-feature fusion is proposed.The proposed algorithm can utilize the different advantagesof several illumination invariants and improve the robust-illumination performance of face recognition under complexlighting environment.The experimental results on the Yale B+and the uncontrolled lighting face database show that theproposed algorithm is effective.
作者
程勇
CHENG Yong(School of Communication Engineering, Nanjing Institute of Technology, Nanjing 211167, China;Fujian Provincial Key Laboratory of Information Processing and Intelligent Control(Minjiang University), Fuzhou 350121, China)
出处
《计算机工程与应用》
CSCD
北大核心
2017年第14期39-44,共6页
Computer Engineering and Applications
基金
江苏省科技计划项目(No.BY2016008-06)
闽江学院福建省信息处理与智能控制重点实验室开放课题资助(No.MJUKF201712)
南京工程学院校级科研基金(No.PTKJ201604)
南京信息工程大学江苏省气象传感网技术工程中心重点实验室开放基金(No.KDXS1503)
关键词
光照不变量
多特征融合
主成分分析
人脸识别
illumination invariants
multi-feature fusion
principal component analysis
face recognition
