Touchless Palmprint Recognition based on 3D Gabor Template and Block Feature Refinement

TL;DR

This paper introduces a large-scale contactless palmprint dataset and a novel 3D convolutional deep learning framework, 3DCPN, that effectively extracts discriminative features for palmprint recognition, achieving state-of-the-art results.

Contribution

It presents the largest contactless palmprint dataset and a novel 3D convolutional network with Gabor feature enhancement for improved recognition accuracy.

Findings

3DCPN outperforms existing methods on multiple datasets.

The large-scale dataset enables robust evaluation of contactless palmprint recognition.

Gabor feature integration enhances curve feature extraction.

Abstract

With the growing demand for hand hygiene and convenience of use, palmprint recognition with touchless manner made a great development recently, providing an effective solution for person identification. Despite many efforts that have been devoted to this area, it is still uncertain about the discriminative ability of the contactless palmprint, especially for large-scale datasets. To tackle the problem, in this paper, we build a large-scale touchless palmprint dataset containing 2334 palms from 1167 individuals. To our best knowledge, it is the largest contactless palmprint image benchmark ever collected with regard to the number of individuals and palms. Besides, we propose a novel deep learning framework for touchless palmprint recognition named 3DCPN (3D Convolution Palmprint recognition Network) which leverages 3D convolution to dynamically integrate multiple Gabor features. In 3DCPN, a novel variant of Gabor filter is embedded into the first layer for enhancement of curve feature extraction. With a well-designed ensemble scheme,low-level 3D features are then convolved to extract high-level features. Finally on the top, we set a region-based loss function to strengthen the discriminative ability of both global and local descriptors. To demonstrate the superiority of our method, extensive experiments are conducted on our dataset and other popular databases TongJi and IITD, where the results show the proposed 3DCPN achieves state-of-the-art or comparable performances.