激光杂志
激光雜誌
격광잡지
LASER JOURNAL
2014年
10期
84-88
,共5页
鲁棒人脸识别%判别性降维%字典学习%线性表示%面部遮挡%户外人脸
魯棒人臉識彆%判彆性降維%字典學習%線性錶示%麵部遮擋%戶外人臉
로봉인검식별%판별성강유%자전학습%선성표시%면부차당%호외인검
Robust face recognition%Discriminative dimensionality reduction%Dictionary learning%Linear repre-sentation%Facial occlusion%Wild face
针对现有的人脸识别算法由于光照、表情、姿态、面部遮挡等变化而严重影响识别性能的问题，提出了基于字典学习优化判别性降维的鲁棒人脸识别算法。首先，利用经典的特征提取算法PCA初始化降维投影矩阵；然后，计算字典和系数，通过联合降维与字典学习使得投影矩阵和字典更好地相互拟合；最后，利用迭代算法输出字典和投影矩阵，并利用经l2-范数正则化的分类器完成人脸的识别。在扩展YaleB、AR及一个户外人脸数据库上的实验验证了本文算法的有效性及鲁棒性，实验结果表明，相比几种线性表示算法，本文算法在处理鲁棒人脸识别时取得了更高的识别率。
針對現有的人臉識彆算法由于光照、錶情、姿態、麵部遮擋等變化而嚴重影響識彆性能的問題，提齣瞭基于字典學習優化判彆性降維的魯棒人臉識彆算法。首先，利用經典的特徵提取算法PCA初始化降維投影矩陣；然後，計算字典和繫數，通過聯閤降維與字典學習使得投影矩陣和字典更好地相互擬閤；最後，利用迭代算法輸齣字典和投影矩陣，併利用經l2-範數正則化的分類器完成人臉的識彆。在擴展YaleB、AR及一箇戶外人臉數據庫上的實驗驗證瞭本文算法的有效性及魯棒性，實驗結果錶明，相比幾種線性錶示算法，本文算法在處理魯棒人臉識彆時取得瞭更高的識彆率。
침대현유적인검식별산법유우광조、표정、자태、면부차당등변화이엄중영향식별성능적문제，제출료기우자전학습우화판별성강유적로봉인검식별산법。수선，이용경전적특정제취산법PCA초시화강유투영구진；연후，계산자전화계수，통과연합강유여자전학습사득투영구진화자전경호지상호의합；최후，이용질대산법수출자전화투영구진，병이용경l2-범수정칙화적분류기완성인검적식별。재확전YaleB、AR급일개호외인검수거고상적실험험증료본문산법적유효성급로봉성，실험결과표명，상비궤충선성표시산법，본문산법재처리로봉인검식별시취득료경고적식별솔。
The performance of face recognition system is seriously impacted by illumination, expression, pose and occlusion variations, for which the algorithm of discriminative dimensionality reduction optimized by dictionary learning is proposed. Firstly, typical feature extraction algorithm PCA is used to initialize dimensionality reduction projection matrix. Then, dictionary and coefficient is computed and the dictionary can match with each other by joint-ing dimension reduction and dictionary learning. Finally, dictionary and projection matrix is outputted by using itera-tive algorithm, and classifier regularized by l2-norm is used to finish face recognition. The effectiveness and reliabili-ty of proposed algorithm has been verified by experiments on extended YaleB, AR and a wild face databases. Experi-mental results show that proposed algorithm has higher recognition accuracy than several other linear represent algo-rithms when dealing with robust face recognition.
