A gradient-based combined method for the computation of fingerprints' orientation field

Estimation of fingerprint orientation fields is an essential module in automatic fingerprint recognition system. Many algorithms based on gradient have been proposed, but their results are unsatisfactory, especially for poor image. In this paper, a gradient-based combined method for the computation of fingerprints' orientation field has been proposed. In our method, we first calculate the first level orientation fields with three different size blocks; and then combine these first level orientation fields together to form the second level orientation field; finally, use the iteration based method to predict orientation. All experiments show that, compared to the prior works, our method is more robust against noise while preserving the accuracy and is capable of predicting. (C) 2008 Elsevier B.V. All rights reserved.