Abstract
In the paper the optimization is applied to determine the threshold of wavelet coefficients in image de-noising aiming at grayscale image corrupted by additive white Gaussian noise and Daubechies wavelet transform. In the method the entropy of an image is the optimality criteria, which makes image de-noising a process of optimizing objective index of an image. The example proves that the presented method has an advantage over the universal soft-thresholding method in both the signal-to-noise ratio and the entropy.