Palmprint Recognition Based on Modular PCA and LS-SVM
Kunlun Li, Yaxin Zhang, HuanHuan Liu, Shuo Sun · 2015
In this paper, in terms of feature extraction, according to the characteristics palmprint image we improve PCA algorithm and design modular PCA algorithms that is suitable for palmprint image.The whole palmprint image is divided into a plurality of sub-block images, then use principal component analysis.The improved feature extraction method is verified by the experiment and it can improve the recognition accuracy rate, especially in the case of single sample palmprint recognition.In the aspect of feature recognition, we train LS-SVM classifier to complete the palmprint recognition system.