Dorsal Hand Vein Recognition Using Gabor Feature-Based 2-Directional 2-Dimensional Principal Component Analysis

摘要

Dorsal hand vein recognition is an emerging biometric techniques researched today. In this paper, we propose a novel approach, the Gabor magnitude feature-based 2-directional 2-dimensional principal component analysis (2D)~2 PCA, for dorsal hand vein recognition. The characteristic of the approach is to combine local and global information to classify. In this approach, Gabor filter bank are used to extract the Gabor image of different frequencies and orientations from the region of interest (ROI) in the dorsal hand vein images. We sum all magnitude from Gabor images of different frequencies and orientations to represent a Gabor magnitude feature (GMF). Then, (2D)~2 PCA is applied for dimensionality reduction of the feature space in both row and column directions and the subspace is classified by minimum distance classifier (MDC). This method is not only robust to illumination and translation, but also efficient in feature matching. The experiment results on our large dorsal hand vein database show that high accuracies (99%) have been obtained by our proposed approach and is feasible for dorsal hand vein recognition.