Computation on fingerprint orientation field based on direction weight diffusion

To solve the problem of low reliability of orientation filed computation in the poor quality fingerprint block, a method for computing the orientation field based on the diffusion of the direction weights was proposed. Firstly, the square direction gradient method was applied to extract the rough direction of the fingerprint field, and then the quality of the fingerprint was evaluated by using the gradient consistency based on the weights. The quality block was divided into four levels. Subsequently the neighborhood discrimination method was performed to correct the misjudged fingerprint block. After correcting the misjudged quality block, the number of the evaluated blocks in the neighboring block was counted. Thereafter, the blocks with the same quality score were divided into three priority levels. Finally, the fingerprint orientation field was iteratively calculated according to the quality and priority. During the iteration process, different weight values were assigned according to the square gradient of the neighborhoods of the block to be evaluated. After the iteration of the evaluation block was completed, the final orientation field was obtained. Experimental results show that the proposed method can improve the reliability of the low-quality region through quality diffusion, thereby enhancing the accuracy of the regions with poor reliability. © 2019, Editorial Board of Journal of Huazhong University of Science and Technology. All right reserved.