Optimizing Parameters of Discrete Wavelet Transform for Noise Reduction in Fingerprint Images
Indrit Enesi, Ledion Liço · 2021
Fingerprint recognition is an important way in person's identification. Noise reduction in fingerprint images increases the accuracy of minutiae detection. Wavelet transforms are widely used in noise reduction in image enhancement. Discreet Wavelet Transform is a method widely used in this field. It is important to find the appropriate filters and mother wavelet. Setting the appropriate parameters has impact in getting optimal results. The parameters analyzed here are the number of decomposition levels and the size of the filter window. The metric used to measure noise reduction is PSNR.