Hand-dorsa Vein Recognition Based on Improved Partition Local Binary Patterns

摘要

In this paper, a new feature descriptor is presented and proposed for personal verification based on near infrared images of hand-dorsa veins. This new feature descriptor is a modification of the previously proposed partition local binary patterns (PLBP) by adding feature weighting, combining multi-scale PLBP and fusion with structure information. While addition of feature weighting aims to reduce the influence of insignificant local binary patterns, combination of multi-scale features aims to get more texture information and fusion with structure feature aims to increase binary information. Testing on a large database with more than two thousand hand-dorsa vein images, Multi-scale PLBP (MPLBP) is shown to be more effective than the original PLBP and Weighted PLBP (WPLBP), and offers a better performance in recognition of hand-dorsa vein images with a correct recognition rate reaching approximately 99% using a simple nearest neighbor (NN) classifier.