Fingerprint Image Enhancement: Algorithm and Performance Evaluation
IEEE Transactions on Pattern Analysis and Machine Intelligence
Hi-index | 0.00 |
In order to reserve more texture features of fingerprint while denoising, a new fingerprint enhancement method based on wavelet transform and unsharp masking is proposed in this paper, which uses multiresolution analysis and local time-frequency analysis characteristics of wavelet. First of all, a fingerprint image is decomposed by wavelet analysis. And then some different treatments are made for the different frequency graphs according to their different characteristics. Third, the initial enhanced fingerprint image is reconstructed by the wavelet coefficients which have been adjusted by before. Finally, unsharp masking is used to process the fingerprint image to enhance the details more. Experimental results show that this method can filter out the noise and enhance the texture features, and thus improve the qualities of the images.