Development and evaluation of the authentication systems by using phase-only correlation palm print identificaton methods
Osman Nuri Uçan, Oğuz Bayat, Melih Burak Coskun · 2017
One of the most important issues in our world of technology is information security. In this process, many biometric security systems and algorithms have been developed using physical features. This paper will investigate palm prints used in biometric authentication and matching systems. The aim is to match two images of one hand of the person taken from the different perspectives and damaged because of the action. The image taken from the palm print reader is softened through the Gauss filter. Our reference area that has the characteristics of this image has been separated from the image as a specific and fixed area by certain algorithms. In this separation process, certain difficulties arose because of the image which could not been taken from the same perspective, differences in the sizes and differences in the sections. These images have been compared in different systems after certain trials and the best solution has been aimed. Phase correlation function has been used as the measure of the similarity between two images. Sub-pixel corresponding matching method has been used for matching of the images in orientational pixel level. Data set was taken from CASIA database. Contribution to palm authentication systems by means of the systems developed is aimed.