A novel face recognition system based on combining eigenfaces with fisher faces using wavelets
Year: 2010
Type: conference-paper
Abstract: In this paper, a biometric system is proposed based on facial features. This proposed system uses an appearance based face recognition method called 2FNN (Two-Feature Neural Network). PCA and LDA are two different feature extraction algorithms used to extract facial features, and then these extracted features are combined using wavelet fusion. The proposed system uses neural networks to classify facial features. Major modules of the proposed system are: extract images from the database; preprocess the extracted images; feature extraction using PCA; feature extraction using LDA; wavelet fusion of the extracted features from PCA and LDA; and neural network based classification. Features are extracted using both PCA and LDA to improve capability of LDA when few samples of images are available. Wavelet fusion and neural networks are used to improve the classification accuracy. The proposed system shows improvements over the existing methods. Preliminary experimental results have shown high accuracy of the system in terms of the correct recognition rate (98.50%) and the equal error rate (1.50%).
Source: Procedia Computer Science
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W2033198680