Palmprint Recognition Based on Two-Dimensional Matched Filter

As one of the palmprint recognition algorithms,the Gabor filter and orientation code based method has obtained widely researches and applications.However,this class of method is easily influenced by noise and defocus.Aiming at the above shortcoming,this paper proposes a 2D matched filter based method,which firstly uses the match filter to extract features and then,employs the double-orientations code (DOC) scheme to encode the result of filter and measure the similarity of palmprint.The experimental results show that the proposed method is robust against noise,defocus and translation,which verifies the effectiveness of the proposed method.