Infrared face recognition based on DWT and BP neural network

Abstract

An efficient infrared face recognition method by combining discrete wavelet transformation(DWT) and BP neural network was proposed. Firstly, to get the principle information and reduce the dimensions of face, each infrared face image was decomposed into DWT coefficients by using two-level DWT. Then, the DWT coefficients was used to extract the useful features. Finally, based on the useful features extracted by DWT coefficients, the recognition result was getten by training of the BP neural network. The experiments conducted illustrate that the method proposed in this paper has better performance.