Design And Implementation Of Face Recognition System

The face recognition is the use of comparative analysis human visual characteristic information for identification of computer technology,including face detection, location and recognition. as a kind of biometric recognition technology, due to its direct, convenient and friendly characteristics, face recognition technology get more and more attentions, become a very active field of pattern recognition and artificial intelligence research directions.This thesis studied and implemented a face recognition system based on static images. The system includes a face image acquisition, image preprocessing, face detection and location, facial feature extraction and recognition of several parts. In the image acquisition, the face image through DirectShow camera image acquisition.In the Image preprocessing, we chose the method of gray transform. In the face detection positioning portion,the image training and detection of human faces and separated is by Adaboost algorithm.In the feature extraction and recognition section, Because of support vector machine (SVM) is suitable for dealing with small sample, high dimension and generalization performance, etc.First we use principal component analysis (PCA) algorithm for face image feature extraction, and then using rough set to reduce the dimensions of the facial featureand, at last we use SVM for classification and identification.In this paper, the design of the system is based on the Visual Studio2010development environment and machine vision library OpenCV.