TW 264

M. Jansen, G. Uytterhoeven and A. Bultheel
Image de-noising by integer wavelet transforms and generalized cross validation


De-noising algorithms based on wavelet thresholding replace small wavelet coefficients by zero and keep or shrink the coefficients with absolute value above the threshold. The optimal threshold minimizes the error of the result as compared to the unknown, exact data. To estimate this optimal threshold, we use Generalized Cross Validation. This procedure does not require an estimate for the noise energy. This paper illustrates the method for wavelet transforms that map integer grey-scale pixel values to integer wavelet coefficients.

report.pdf / mailto: M. Jansen