The Intelligent Security Monitoring System Based On Face Recognition

With the improvement of living standard and the rapid development of socialeconomy of our country, people pay more and more attention to the security and property.The traditional video monitoring system needs to be equipped with professional securityofficer. Not only wasting time, but also the security level is low. To Design and develop anew type of intelligent security monitoring system is very necessary. Aiming at the villausers, man designs a kind of intelligent security monitoring system. It has abnormalbehavior detection, face recognition, GPRS MMS send functions, including videocollection equipment selection, the choice of GPRS modem and the realization of the facerecognition algorithm.General face recognition is divided into three steps: face detection, face featureextraction and face recognition. This paper is mainly about principal component analysisintroduced. PCA is a kind of linear methods. It is the optimal method to dimension ofcompression. The use of PCA to face sample data transformation of the low dimensionaldata in all compression algorithm contains the most abundant information. PCA methodanalyzes the data covariance array, it can effectively eliminate the correlation betweeneach data. This method is based on solving the characteristic equation, selection of thecharacteristic value for larger the eigenvectors corresponding to the characteristics ofmatrix of face image transformation. PCA is easily by sample illumination, poseinformation influenced. The analytic method is in this paper improved. Instead of PCAmethod is LBP method to be used. LBP operator focuses on describing the sample texturefeature. On the other hand, LPB can better solve the problem, which man in the past had,for example: position, illumination, pose, expression and so on. The traditional PCA method and LBP method was used in this paper. The two methodwas used to do the experiment of face recognition on the Yale face library. The statisticsfor the recognition time and recognition rate for each algorithm and the comparisonbetween them was made. Some image processing function of Visual C++6.0and theOpenCV library was used to realise the face recognition method of the LBP and PCA, andit had a good effect.