Directional Field Computation for Fingerprints Based on the Principal Component Analysis of Local Gradients

This paper discusses a new method, based on the principal component analysis, to estimate the directional field of fingerprints. The method not only computes the direction in any pixel location, but its coherence as well. It will be proven that this method provides exactly the same results as the method that is known from literature. Undoubtedly, the existence of a completely different solution method increases the insight into the problem’s nature. Segmentation of fingerprints is discussed to illustrate the application of the coherence of the directional field. Keywords— image processing, fingerprint recognition, orientation estimation, principal component analysis.