A Novel User-Specific Palmprint Verification Approach

For the stable and unique characteristics, palmprint is one of the relatively new biometrics. In this paper, we propose a new user-specific method, which selects user-specific palmprint blocks by several feature extracting methods considering the stability of the selected blocks. First, we use a sliding block while extracting every part´s features by different methods. Secondly, the most common classify method, the k near neighborhood method, is used. Finally, we will choose the specific block and dimension of extracted features corresponding to the best verification rate for every user. For proving the effectiveness and feasibility, we compare the proposed specific feature selection approach with the methods without blocks selection on a database. The experimental results show the proposed approach can achieve higher verification rate compared with the common methods, and more stable in palmprint verification.