Fingerprint pore matching method using polar histogram

Abstract

Pores on the fingerprints have proven to be useful features for fingerprint recognition and successfully used in Automatic Fingerprint Recognition Systems (AFRS). Several methods for pore matching have been reported and these methods employ minutiae or ridge information. This paper presents a pore matching method using relative locations between extracted pore centers. The proposed pore matching method first generates a descriptor for each pore using polar histogram and establishes the correspondences between the pores from the reference image and the target image. Then, RANSAC (RANdom SAmple Consensus) algorithm is used to refine the correspondences and calculates the final matching score. A series of experiments show that fingerprint recognition could be achieved using only relative locations between the pores.