Application of reversible denoising and lifting steps to DWT in lossless JPEG 2000 for improved bitrates

Skipping Selected Steps of DWT Computation in Lossless JPEG 2000 for Improved Bitrates

“…In [3], motivated by the observation that the lifting step of a color space transform may increase the amount of noise in an image component, we applied to the DWT the reversible denoising and lifting steps (RDLS) [4], which are lifting steps integrated with denoising filters. In brief, we replaced the prediction (Eq 1) and update (Eq 2) lifting steps with the RDLS-modified counterparts presented in below Eq 3 and Eq 4, respectively: where s d i is the denoised sample s i obtained using a deterministic denoising filter.…”

“…Despite the inherently lossy nature of denoising, RDLS exploiting denoising and transforms consisting of several such steps are perfectly invertible. For further details and RDLS examples the reader is referred to [3,4,13].…”

“…The basic heuristic is based on one used in [3] for the selection of denoising filters. It consists of the greedy steps described below, in which step B may be repeated for a certain number of iterations.…”

“…The heuristic used with None and Null filters results in exactly the same transform, as obtained with BH. Obviously, by using it without the Null filter we get RDLS-DWT exactly as in [3]. By using None, Null, and the regular denoising filters we combine SS-DWT with RDLS-DWT; the heuristic for each lifting step may decide to keep it unchanged, skip it, or apply to it RDLS with an actual denoising filter.…”

Application of reversible denoising and lifting steps with step skipping to color space transforms for improved lossless compression

“…In Ref. 11, RDLS was applied to discrete wavelet transform (DWT) in lossless JPEG 2000 compression of grayscale images. The noise filtering was the most effective when applied only to some steps.…”