Consistency analysis on orientation features for fast and accurate palmprint identification

A typical palmprint identification system identifies a query palmprint by matching it with the templates in the database and comparing the similarity score with a pre-defined threshold. However, continual database expansion would make conventional brute force search method inefficient for large-scale palmprint identification. In this paper, we propose a consistent orientation pattern (COP) hashing method to enforce fast search. Using the orientation and response features extracted by steerable filter, we first give an analysis on the consistency of orientation features, and then introduce a method to construct COP using the consistent features. Because the COP is very stable across the samples of the same subject, the COP hashing method can find the target template quickly and thus can lead to early termination of the searching process. Experiments have been carried out on several real and synthetic databases, and the results show that the COP hashing method can accelerate the identification process by more than an order of magnitude, and can achieve comparable identification accuracy while compared with the state-of-the-art approaches.