Minutiae and corner detection in fingerprints without image enhancement for real time recognition

Abstract

Most fingerprint verification and recognition systems are based on the minutiae feature points. This paper presents a novel and robust edge based minutiae and corner detection algorithm designed for fingerprints with very small time consumption. This detector extracts the feature points on ridge edges without image enhancement and thinning. In turns, it provides an important time reduction which is required for real time and automatic fingerprint recognition systems. In addition, the detector localizes the corners on the edges of the ridges which characterizes the ridge curvature and its presence corresponds to a deviation on the ridge direction. The combination Minutiae/Corners forms a new set of features that will increase the performance of fingerprint recognition system. Finally, the results achieved are compared with those obtained by other techniques in the literature. Results show that the proposed method performs much better in terms of time efficiency with similar Equal Error Rate (EER) performance measure.