Non-uniform threshold as an alternative to uniform threshold in denoising in wavelet domain

In this paper we present the advantage of non-uniform over uniform threshold wavelet shrinkage denoising method, applied on noisy signals with signal dependent noise. We illustrate our results by comparing the noise energy after using the both filtration methods on the same set of artificially noise contaminated images. The experiments are made with NPR-QMF filter banks instead with the filter banks that are commonly used in wavelet applications.