An Improved Face Recognition Based on ICA and WT

A face recognition method based on independent component analysis and wavelet transform is proposed. Firstly an image is decomposed using WT into different frequency sub-bands, and then ICA is applied on wavelet sub-bands to get the independent vector, which includes the main information of original image, finally face recognition is implemented with the subspace comprised by these basis vectors. We compared our methods with two face recognition algorithms, ICA and WT. In the experiments, the nearest-neighbor classifier is used to recognize different faces from the ORL face database. Experimental results show that the proposed method improved the recognition rate effectively, the best accuracy rate can reach 92%.