Title :
Face recognition for target detection on PCA features with outlier information
Author :
Chen, Yen-Lun ; Zheng, Yuan F.
Author_Institution :
Ohio State Univ., Columbus
Abstract :
This work addresses face recognition for detecting a small and particular set of individuals over a huge population of people. A new approach is developed which uses training samples of both target classes and non-target outliers. Two-stage principal component analysis (PCA) schemes are proposed for flexible feature extraction. Experimental results reveal that the new approach improves the accurate rate of detection significantly.
Keywords :
face recognition; feature extraction; principal component analysis; target tracking; face recognition; flexible feature extraction; outlier information; principal component analysis; target detection; Computer vision; Covariance matrix; Eigenvalues and eigenfunctions; Face detection; Face recognition; Feature extraction; Object detection; Principal component analysis; Symmetric matrices; Terrorism;
Conference_Titel :
Circuits and Systems, 2007. MWSCAS 2007. 50th Midwest Symposium on
Conference_Location :
Montreal, Que.
Print_ISBN :
978-1-4244-1175-7
Electronic_ISBN :
1548-3746
DOI :
10.1109/MWSCAS.2007.4488700
