Improved CNN-Segmentation-Based Finger Vein Recognition Using Automatically Generated and Fused Training Labels

Abstract

We utilise segmentation-oriented CNNs to extract vein patterns from near-infrared finger imagery and use them as the actual vein features in biometric finger vein recognition. As the process to manually generate ground-truth labels required to train the networks is extremely time-consuming and error prone, we propose several models to automatically generate training data, eliminating the needs for manually annotated labels. Furthermore, we investigate label fusion between such labels and manually generated labels. Based on our experiments, the proposed methods are also able to improve the recognition performance of CNN-network-based feature extraction up to different extents.