A systematic gradient-based method for the computation of fingerprint’s orientation field

Orientation field which describes the local direction of ridge-valley pattern, play a very important role in automatic fingerprint recognition. In this paper, a systematic gradient-based method for the computation of fingerprint’s orientation field is proposed. This method mainly includes two procedures: the coarse level orientation computing, and the orientation predicting. In the first procedure, the effective point gradient vector and the composite block are proposed for overcoming the limitation caused by normalization, and the contradiction between accuracy and robustness respectively; in the second procedure, an iteration based predicting method is adopted for improving the robustness against noise. All experiments show that: compared with the previously proposed gradient-based approaches, our method does not only possess the advantage of high accuracy, but also be more robust against noise and be capable of predicting.

Graphical abstractFigure optionsDownload full-size imageDownload as PowerPoint slideHighlights► A systematic gradient-based method for the computation of fingerprint’s orientation field (OF). ► This method mainly includes two procedures: the coarse level OF computing, the OF predicting. ► The effective point gradient vector is proposed for overcoming the shortcomings caused by normalization and non-normalization. ► The composite block is proposed for overcoming the contradiction between accuracy and robustness. ► For improving the robustness against noise, the iteration-based method is employed to predict the orientation of noise region.