Face Recognition System Approach Based on Neural Networks and Discrete Wavelet Transform

The technology of face recognition is attractive and full of technology research challenges; It is used for recognizing people by using digital images. Although face recognition has an important role in several areas such as security, face recognition technology still encounters many challenges that need to be solved with more scientific methods. One of These challenges lead can be the variations of the face of the same person due to lighting or pose. This project explores and investigates the use of combined hybrid algorithms based on neural networks and discreet wavelet transform for face recognition in order to enhance the recognition rate for a face from identified data set of faces. Two techniques have been used in this research; the First one is applying the discrete wavelet transformation method in order to improve and compress the images of the data set. The second one is implementing a well-known approach called Principal Component Analysis. The training and testing face images are selected from ORL database, which contains 400 images for 40 different persons and have minimum pose variation. The experimental results confirmed that the proposed methodology provides a feasible and effective solution for recognizing faces.