Face Recognition with Neural Networks

Biometric is an emerging trend of science for identifying a person based on physical and behavioural attributes. Large scale of biometric systems are used both in commercial and government agencies to increase the level of security. Face recognition is the powerful method of recognizing an individual. It is a computer application which compares the facial features with the database image. A unique approach is introduced for recognizing multi pose images with different conditions. The classification is performed by using Back Propagation Algorithm and Radial Basis Algorithm and Adaptive Neuro Fuzzy Inference System. The proposed method is based on extraction and Eigen face method. Experimental results are tested using FERET database and shows that the proposed method is effective for face recognition. The feature extraction using PCA is to reduce the dimensionality and to calculate the feature points. The performance evaluations between the proposed methods are analyzed. The characteristic features calculated using PCA called Eigen faces are obtained from the stored images are combined with the Back Propagation Neural Network, Radial Basis Function Network and Adaptive Neuro Fuzzy Inference System algorithms for recognition of images. For extracting the feature points from the input image, entropy and mean of each image are detected. The experimental results on performance of testing images are compared with Back Propagation, Radial Basis Function Network and Adaptive Neuro Fuzzy Inference System algorithms. Thus Face recognition can be achieved through the architectures and algorithms of artificial neural networks.