Palmprint recognition with 2DPCA+PCA based on modular neural networks

Abstract

In this letter, a novel modular neural network (MNN) classifier, which partitions a K -class problem into many much easier two-class problems in sub-subspaces, was proposed to perform palmprint recognition. Moreover, in order to make palmprint recognition more accurate, we introduced 2DPCA technique into the extraction of palmprint features, and removed the illumination information from the collected palm images using w/o3 technique. Our approach was compared with several existing methods, and obtained a satisfying classification performance on the Hong Kong Polytechnic University (PolyU) Palmprint Database.