An Improved Fingerprint Singular Point Detection Algorithm Based on Continuous Orientation Field

Abstract

It is very important to detect singular points (core and delta) accurately and reliably for classification and matching of fingerprint. In this paper, an improved method for singularity detection in fingerprint images, which based on continuous orientation field, is proposed to improve accuracy of the position and reliability of the singularity. Firstly, the blocks which may contain singularities are detected by computing the Poincare Index. Then, the singularities are detected in the block images. Experiment show that the proposed method can overcome the shortcoming of the traditional method to great extend and is robust to poor quality images.