Tuning the locality of filtering with a spatially weighted implementation of random spray Retinex

Abstract

The human color sensation depends on the spatial distribution of the colors in the viewed scene. This principle is at the basis of the random spray Retinex (RSR) algorithm. In this work, we modify RSR by integrating its approach with a method to weight and tune the locality of spatial image information. This modification allows for spatial control of the local effect of RSR on image color filtering. We study the performances of this spatially weighted version of RSR on a public image dataset by analyzing and comparing several image features of the output image and its local properties.

Point‐based spatial colour sampling in Milano‐Retinex: a survey

“…Two RSR variants, i.e. Light-RSR [14] and swRSR [15]. Light-RSR basically introduces Gaussian spatial and intensity smoothing to further decrease the chromatic noise and to speed up the execution time.…”

“…The swRSR [15] was mainly devised to investigate and tune the locality of the spatial colour filtering of RSR. swRSR adopts the random sampling scheme of RSR, while it modifies the lightness computation by introducing a function w with weights the spray pixels depending on their spatial distance from the spray centre.…”

“…The remaining part of the paper is organised as follows: Section 2 explains the differences between the Milano-Retinex family and the original computation. Sections 3-6 describe, respectively, RSR [13] and its variant Light-RSR [14], spatially weighted (swRSR) [15], and the exact population-based formulation RSR-P [16]. Section 7 describes the probabilistic approach quantile-based approach to RetIneX (QBRIX) [17], followed by spatio-temporal Retinex-like envelope with stochastic sampling (STRESS) [18] and spatio-temporal Retinex-like envelope with total variation (STRETV) [19] in Section 8 and T-Rex [20] in Section 9.…”

Milano Retinex family

“…The same approach is exploited by swRSR and ETR. In swRSR (spatially weighted RSR), 34 the spray sampling is weighted by a figure of distance. In ETR (energy-driven path search for TR), 35 the paths are computed as local minima of an energy functional, depending on the image gradient.…”

Designator Retinex, Milano Retinex and the locality issue

“…In swRSR (spatially weighted RSR) [19], the spray sampling is weighted by a figure of distance. In ETR (Energy-driven path search for Termite Retinex) [20], path generation is realized solving partial difference equations (PDE) on image gradients.…”