Face Recognition Based On PCA And2-D Gabor Wavelet Transform

In recent years, with the rapid development of computer technology and the peoplepaying more attention on information security, the original identity authentication has beenunable to meet the needs of social reality, so biometric identification technology is coming.Compared with fingerprint recognition and other traditional means of identification, facerecognition technology is easier to be accepted by users because of its accuracy, concealmentand non-intrusive features. Just for this reason, face recognition technology has a wide rangeof applications in many fields. In recent years, facial feature extraction in face recognitiontechnology has become one of hot spot based on biological characteristics.Based on summary of the content and method of face recognition technology, this paperdiscusses in detail two types of feature extraction method based on subspace analysis: PCA(principal component analysis) method, and2-D Gabor wavelet transform. Then we elaboratethe main idea of the two methods in detail and introduce their algorithm process andimplementation. In this paper, firstly the original Adaboost algorithm has been improved andapplied to face detection. PCA has a big advantage on expression of global features of face andGabor wavelet has a big advantage on expression of local features of face. By making use ofthe advantages, we extract global features of face by PCA and extract local features of face by2-D Gabor wavelet, then we establish a double-classifier (global classifier and unifiedclassifier) by combining the two features together. Finally, the research of face recognitiontechnology based on PCA and2D-Gabor wavelet transform is simulated on the ORL facedatabase and self-built face database. Experimental results show that the improved algorithmcan not only effectively improve the human face recognition rate and recognition speed, butalso reduce the false recognition rate, and it still maintains high recognition rate, more than90%, in short, the algorithm has a good practical value.