Study On Methods Of Color Image Filtering Based On Visual Sensitivity

To eliminate the influence of visual difference for color image filtering,methods to compute the visual sensitivity value of three primary colors and image evaluation ware given.A series of filtering algorithm ware proposed by used visual sensitivity value into color image filtering.①Multiple the advantages of switching filtering algorithm and weighted filtering algorithm,we proposed Visual sensitivity Rough set Directional Distance Filter(VRDDF). This filter algorithm detected the nose by rough set theory first,then used visual sensitivity to correct the test results,that can reduce the false detection rate effectively.In order to improve the efficiency of the filtering,this paper made reasonable weighted to Directional Distance Filter(DDF) based on the space distance perception function.Then use it to filter noise pixels only,and keep the value of pixels which aren’t polluted,that can prevent noise transmission.Experiment results show that VRDDF is better than existing filtering algorithm in color keeping,noise filtering rate,especially in edge detail preservation;②Comprehensive consideration the effect and the operation time of filtering, put forward the adaptive detection window and minimum minimal filter window filtering algorithm,Experiment results show that it is better than existing filtering algorithm in color keeping,noise filtering rate and edge detail preservation.And under the various density noise, it has the least the normalized mean square, and the most peak signal to noise ratio.③In this paper, a filtering method for boundary noise is proposed,which treats boundary and internal noises separately, therefore reduces the interactive influences between each other, this method can filter the boundary noise effectively.All the filtering algorithm above have good filter ability, they are good at edge detail preservation and color keeping.