Efficient Wavelet Entropy-Based Face Recognition System

In the presented paper, face images modeling in different circumstances such as occlusion, poses, and illumination was investigated. A novel research based on a new combination of discrete wavelet transform (DWT) packet (WP) methods conjunction with log energy sure entropies is proposed. Probabilistic neural network (PNN) used for classification. The performance proposed method investigated using collection experiments challenges ORL database extended Yale conducted. For testing recognition system database, many classification scores were calculated training/testing sets. comparative results obtained published efficiency performed showed that our superior rate reached 98.5% relatively small, elapsed time about 4.4 sec found. B, averaged at each 42 iterations. Furthermore, sets excellent achieved. contribution study to use entropy task by finding best leads achieving competent form guarantying better results. major this it possible just few real model conditions B.