Palm print ROI extraction using Bresenham line algorithm

Abstract

Hand based biometrie authentication is becoming so popular in the field of information security because it is an accurate and easily accessible procedure to legalize the human identity. Many studies related to palm print recognition have been proposed recently. To achieve superior recognition results, an accurate segmentation of region of interest is very crucial. In this article, a novel palm print ROI extraction algorithm has been presented which extracts a fixed size region from a full hand image. The quadrangular shape ROI covers the maximum possible area over palm and thus consists of more features. The basic coordinate system is modified to find accurate base points at fingers bottom so that a square with longest side can be drawn using Bresenham's line algorithm. The publically available CASIA, PolyU and IIT Delhi Contactless palm print databases have been used for testing. Addressing to the matching problem, a novel Deep-Matching algorithm has been used. The results have been compared with two state-of-art algorithms. It has been observed that proposed algorithm outperforms with EER drop not more than 20%. With this it is clear that the proposed algorithm has been extracting palm ROI more consistently and hence assist to improve the performance of traditional palm print biometric system.