Key points are not available for this paper at this time.
Abstract Fingerprint matching is essential to ensure the robustness of fingerprint identification with respect to the image quality. Usually, fingerprint images are enhanced by one stage in either the spatial domain or the frequency domain. Moreover, fingerprint enhancement needs to be conducted in order to enhance the fingerprint image completely. In recent years, two-dimensional fractional Fourier transform (2D-FRFT) which contains both spatial and frequency information, has been playing a unique and increasingly important role in image processing, the studies on using it for fingerprint image are very rewarding and promising. In this paper, we propose a novel approach which combines the properties of 2D-FRFT and phase correlation technique in order to provide a simple means for estimating the translation and rotation parameters between two fingerprint images in 2D-FRFT domain. Our method relies on 2D-FRFT-based correlation twice: once in the spatial domain based on the average squared difference function (ASDF) to recover the translation and once in the polar domain to estimate the rotation. The effectiveness of the proposed approach is evaluated on the public FVC2004 set a database. Experimental results demonstrate that the proposed method is capable of estimating the translation and rotation in fingerprint images, and its accuracy is comparable to the 2D-FFT-based method.
El Mehdi Ismaili Alaoui (Fri,) studied this question.