The Kind Of Guard System Based On Face Recognition

Biometrics entrance guard system based on the biological characteristics as a condition of recognition of entrance guard system,mainly including(face,DNA,fingerprint,voice,retina,iris,palmprint,etc.)and the face recognition with its non invasive,direct,friendly,natural and convenient wait for a characteristic,become the mainstream in the study of biometrics.Face recognition system is a kind of pattern recognition system based on information processing.It mainly includes two parts:face detection and face recognition.Face detection part is to be found in the whole video frames need to identify the faces,the basic theory is to use the Haar characteristics and integral image,Trained in the face samples and the non-face samples with face classification ability weak classifier,strong classifier,and the cascade cascade classifier,and cascade it to cascade classifier,which is enclosed Adaboost algorithm is deeply touched.Face recognition mainly includes two aspects:feature extraction and pattern classification.Feature extraction part is from the human face image using PCA + LDA extraction can be used to distinguish between different human face image feature information,the role of the second part is to former part extracted feature information classification,the recognition rate of the whole system is determined by the two parts together,so the efficiency of each part about the pros and cons of recognition results.Because the sample camera entrance guard is in oblique high building,so the detected face almost all of the non-positive face,but at present commonly used face perception system can only deal with approximate frontal face.For processing a positive face image,the performance of these systems will fell sharply,so in face recognition,the change of the posture is a hot spot is also a difficulty,so in this paper present,the access system using the lower detection when error detection rate and find relatively the most is face of combination of technology solution.That the combination of two kinds of technology will enhance the safety and efficacy of identify recognition,to ensure optimal identification results,meet the deman,d of the workplace safety and efficient management.