{"title":"The use of threshold LUM smoothers in noised color sequences","authors":"R. Lukac, S. Marchevsky","doi":"10.1109/EURCON.2001.938141","DOIUrl":null,"url":null,"abstract":"This paper deals with impulse noise suppression from noised color sequences by three-dimensional (3D) component-wise threshold LUM (lower-upper-middle) smoothers. The proposed threshold LUM smoothers are distinguished by a best balance between impulse noise suppression and signal-detail preservation, since the smoothing level is adaptively chosen depending on local neighborhood information. In color image filtering, threshold LUM smoothers produce an output with minimal color distortion. Another property of threshold LUM smoothers that must be mentioned is their convergence to qualitative root signals that present a very low deviation from the original. Thus, a filtering process performed by threshold LUM smoothers approaches an optimal situation, where the noise free samples are passed to the output without change whereas corrupted samples are estimated as the most appropriate value from a set of possible outputs of standard LUM smoothers.","PeriodicalId":205662,"journal":{"name":"EUROCON'2001. International Conference on Trends in Communications. Technical Program, Proceedings (Cat. No.01EX439)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2001-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"EUROCON'2001. International Conference on Trends in Communications. Technical Program, Proceedings (Cat. No.01EX439)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/EURCON.2001.938141","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 3
Abstract
This paper deals with impulse noise suppression from noised color sequences by three-dimensional (3D) component-wise threshold LUM (lower-upper-middle) smoothers. The proposed threshold LUM smoothers are distinguished by a best balance between impulse noise suppression and signal-detail preservation, since the smoothing level is adaptively chosen depending on local neighborhood information. In color image filtering, threshold LUM smoothers produce an output with minimal color distortion. Another property of threshold LUM smoothers that must be mentioned is their convergence to qualitative root signals that present a very low deviation from the original. Thus, a filtering process performed by threshold LUM smoothers approaches an optimal situation, where the noise free samples are passed to the output without change whereas corrupted samples are estimated as the most appropriate value from a set of possible outputs of standard LUM smoothers.