Face Recognition System Using SVM Classifier and Feature Extraction by PCA and LDA Combination
Jianke Li, Baojun Zhao, Hui Zhang, Jichao Jiao · 2009
Feature representation and classification are two key steps for face recognition. A novel method for face recognition was presented based on combination of PCA (principal component analysis), LDA (linear discriminate analysis) and SVM (support vector machine). PCA and LDA combination was used for feature extraction and SVM were used for classification. The normalization had been done to eliminate redundant information interference previous to feature extraction. The experiments were implemented on ORL face database with the approach. Compared with PCA and Nearest Neighbor Classifier (NCC) combination method, PCA, LDA and NCC combination method, our approach improved face recognition rate.