Pca Algorithm, Face Recognition System

With the development of society, the request to detect one's identity card (ID) effectively and automatically has been in all respects urgent day by day. Because the biological characteristic is people's inherent attribute, having very strong stability and individual independency, it is the ideal source of information for ID verification. In addition, because the patterns of face are the natural and direct in comparison with other biological characters, the technique of face verification (FV) is widely utilized as the mainstream approach in ID verification.The aim of FV can be easily described as follows: Given a certain static picture or dynamic video picture of scene, try to detect and recognize one or more persons in them on the basis of prearranged face database indicating relevant ID information. In the realm of computer vision, the process of FV consists of three parts: Face Detection (FD), Feature Extraction (FE) and Face Recognition (FR), in which Feature Extraction is the most significant part.FE is a process which transfers the data from primary spaces into feature space, representing them in a lower dimensional space with less effective characters. Up to now, many methods of feature extraction have been proposed. Among them, the algorithm of Eigenface, the most widely-used method of linear map based on PCA (Principle Component Analysis), has become the mainstream criterion to test the performance of various FV systems.However, the traditional algorithm of Eigenface and its transformation has strict restraints on the light in the original image, posing as a hindrance in the practical application. Aiming at this problem, this paper proposes a FR algorithm based on the fusion of algorithms of PCA andLDA. This algorithm utilizes PCA and LDA (Linear Discriminate Analysis) to deal with the picture. Then trains and discerns in BP artificial neuralnetwork. This Algorithm draws and combines together PCA and LDA optimization characteristic got to merge with neural adaptivity of network, have obtained higher discerning rate and fine resisting noise performance.