feed forward neural network based palm pattern classification

Palmprint is one of the most unique and stable biometric characteristics. Although 2D palmprint recognition can achieve high accuracy, the 2D palmprint images can be easily counterfeited and much 3D depth information is lost in the imaging process. Previous works on 3D palmprints has focused on local and global features and also it requires some more training process during the palm print recognition. In this paper, we propose feed forward neural network basedpalm prints which describe shape information and to improve the efficiency ofpalm-print recognition. The feed forward neural network is trained using back propagation algorithm which is being employed to classify the palm images based on its features. Thus our experiment demonstrates that the proposed model leads to higher accuracies and also provides quick results, compared against the pattern recognition technique that uses SVM classifier for quantification.