Description |
PURE-LET has been recently proposed [1] as noise removal strategy from Poisson-count images. Specifically, PURE (Poisson unbiased risk estimator) is an unbiased estimate, defined in the Haar wavelet domain, of the mean-squared error between the original image and the estimated image. PURE-LET attempts to estimate the true image from the noisy image by minimizing PURE. PURE-LET involves inversion of a small size matrix.
[1] F. Luisier, C. Vonesch, T. Blu, M. Unser, "Fast Interscale Wavelet Denoising of Poisson-corrupted Images", Signal Processing, vol. 90, no. 2, pp. 415-427, February 2010. |