Study of Face Recognition Algorithm Based on Proximal Support Vector Machine

Face recognition algorithm based on support vector machines (SVM) have better recognition rate, but the time of training is very long when it have a large number of sample. To overcome this shortcoming, in this paper, the face recognition algorithm based on the proximal SVM (PSVM) was proposed, which the first face image through principal component analysis (PCA) for dimensionality reduction and then use PSVM to classify. The experimental results in ORL and Yale face database show that the training time had a greater reduction and the recognition rate slight lower than the traditional SVM. The reduction of training time is SVM´s a few percent. In particular, it has better improve of training time when its dimension is not high and have a larger number of samples.