A Gradient-Based Robust Method for Estimation of Fingerprint Orientation Field

Automatic fingerprint identification system (AFIS), which is one of the most important biometric authentication, has been extensively studied and good performance on small database is obtained, but there still exist some critical issues such as long processing time on large databases and low matching rate on poor image. To solve these problems, improvements of fingerprint classification and identification are needed. As a global feature of fingerprint, orientation field which describes the local direction of the ridge-valley pattern plays a very important role in both topics mentioned above. Many fingerprint orientation estimating methods based on gradient have been proposed, but their results are not very satisfactory, especially for poor images. In this paper, a gradient based robust method for estimation of fingerprint orientation fields is proposed. This new method mainly comprises three steps: firstly, normalize the point-gradient vectors and calculate the block-gradient vectors and the corresponding block-coherence; then detect the noisy areas to eliminate the side effect of noise diffusing; finally, re-estimate all block-gradient vectors based on iteration and transform the gradient field to orientation field. Compared with the previously proposed gradient-based methods, experiments conducted on FVC 2000 and FVC 2004 show that the proposed method is more accurate and more robust against noise, and is able to predict orientation field within the large noisy areas.