Hand Side Recognition and Authentication System based on Deep Convolutional Neural Networks

Abstract

The human hand has been considered a promising\ncomponent for biometric-based identification and authentication\nsystems for many decades. In this paper, hand side recognition\nframework is proposed based on deep learning and biometric\nauthentication using the hashing method. The proposed approach\nperforms in three phases: (a) hand image segmentation and\nenhancement by morphological filtering, automatic thresholding,\nand active contour deformation, (b) hand side recognition based\non deep Convolutional Neural Networks (CNN), and (c) biometric\nauthentication based on the hashing method. The proposed\nframework is evaluated using a very large hand dataset, which\nconsists of 11076 hand images, including left/ right and dorsal/\npalm hand images for 190 persons. Finally, the experimental\nresults show the efficiency of the proposed framework in both\ndorsal-palm and left-right recognition with an average accuracy\nof 96.24 and 98.26, respectively, using a completely automated\ncomputer program.