Denoising Fingerprint Image Using An Improved Wavelet Threshold Function Method

Fingerprint recognition technology is often disturbed by noise. In the process of using wavelet threshold denoising method to reduce fingerprint image noise, the selection of the threshold function is particularly significant. Therefore, a new threshold function with adjustable parameters, which is between traditional soft and hard threshold functions, is constructed. The new threshold function overcomes the discontinuity of the hard threshold function at the threshold point, improves the constant deviation of the soft threshold function and can reach zero deviation under certain conditions. Simulation experiments show that using the new threshold function to remove salt-and-pepper noise and Gaussian noise in fingerprint images during the wavelet threshold denoising process results in lower root mean square error and degree of distortion compared to using traditional and improved threshold functions, while increasing peak signal to noise ratio, correlation coefficient, and structural similarity index measure. This not only provides a method to improve the threshold function, but also offers an effective means for improving the denoising quality of fingerprint images.