Based On Principal Component Analysis Of Face Recognition Algorithm

Face recognition technology is an important computer application technology, content, pattern recognition, artificial intelligence, one of the key research topics. The difficulties of face recognition is that a rich person's face, while the increase with age, the overall situation will face some changes, in the hair, glasses and some ornaments of the block, the different angles that the light will make a number of enhanced features to varying degrees and watered down, to bring the difficulty of face recognition a great challenge.This according to statistical theory, the multivariate statistical analysis of principal component analysis to face recognition algorithm is introduced to, from the perspective of statistical pattern recognition to study the human face image feature extraction algorithm using principal component analysis to get a matrix the recognition of high dimensional data space to low dimensional space, so linear, this matrix can obtain the covariance matrix of feature vectors obtained, without parameters, greatly simplifying the calculation.In this paper, the British ORL face database as the basis for experimental analysis. ORL face database is the most widely used standard database, which contains a large number of comparisons. The database consists of a different time, a black background and details of facial expression changes were composed of 400 gray image. Experimental results show that the algorithm of recognition accuracy and recognition speed has improved more, the correct rate of 90%.