Hierarchical Palmprint Coding with Multi-Features

Summary

Palmprint is regarded as one of the most unique, reliable and stable personal characteristics and palmprint verification provides a powerful means to authenticate individuals for many security systems. Palmprint feature extraction and matching are two key issues in palmprint identification and verification. In contrast to the traditional techniques, we propose a hierarchical palmprint coding scheme to integrate multiple palmprint features for guided palmprint matching. The combination of four-level features including Level 1: global geometry feature; Level 2: global texture energy; Level 3: fuzzy “interest” line; and Level 4: local texture feature possesses a large variance between different classes while remaining high compactness within the class. The coarse-level classification by Level 1 and Level 2 features is effective and essential to reduce the number of samples significantly for further processing at fine level. Then, Level 3 and Level 4 local texture leads to fast search for the best match. The experimental results provide the basis for the further development of a fully automated palmprint-based security system with high performance in terms of effectiveness, accuracy, robustness and efficiency.