2015 International Symposium of Computer Application and Information Technology (IsCait 2015)

330 / 1971-01-01 00:00:00

Face Recognition Based On Adaptive Discriminative Local Patch Alignment

face recognition,Adaptive local discrimination analysis,Local patch alignment,Feature extraction

Abstract—Due to the influence of external environment, the number of samples and the disturbance conditions in the process of face recognition, the error rate of classifiers will increase dramatically which results in the mismatching disaster situation. To solve this problem, we adopt a modified perceptron and pre-recognition algorithm improve and solve this problem on the basis of face recognition without closed and rejection set, After several tests made on real-time face recognition system designed on VC++ platform, the results show that the classification algorithm designed in this paper has good adaptability and robustness, low mismatching rate, fast speed and is convenient for practical application.