New algorithm of extraction of palmprint region of interest (ROI)

Abstract

Palmprint contains numerous patterns that are unique and distinct in detecting human identity. The process of extracting the palmprint ROI is very important since its result provide initial information for personal identification. In obtaining palmprint ROI, there are numerous algorithms that have been proposed by past researchers. However, as best to our knowledge, the extracted ROI using these algorithms only cover a small part of the palmprint. Due to this limitation, some important features that can be used as an individual identification might be neglected. In this research, a new algorithm for extracting the palmprint ROI is proposed. The performance of the proposed algorithm is compared to two other existing algorithms. The three algorithms are tested based on the location and size of the extracted ROI. The result obtained shows that the proposed algorithm has successfully extract larger palmprint ROI compared to existing algorithms. The results provide platform for further use on identifying and verifying the identity of an individual.