D. Prakash, Pramati Kalwad, V. Peddigari, P. Srinivasa
{"title":"Automatic reflection removal using reflective layer image information","authors":"D. Prakash, Pramati Kalwad, V. Peddigari, P. Srinivasa","doi":"10.1109/ICCE.2015.7066520","DOIUrl":null,"url":null,"abstract":"This paper tries to address the problem of removing unwanted reflection layer from the image mixtures. These reflections may occur due to semi-reflective glass mediums. This demands separating the input image into the reflecting layer and the subject layer, also known as, background layer which is the actual scene itself. But decomposing a single input image into these two layers is a massively ill-posed problem with infinite combinations of decomposition. This type of problem falls under the category of blind source separation. There are ample classical separation approaches available in the literature which either requires multiple images or works on a single image with user assistance. In this paper we propose a method for separating the two layers from a single input image without any user or human intervention using some prior information about the reflective layer.","PeriodicalId":169402,"journal":{"name":"2015 IEEE International Conference on Consumer Electronics (ICCE)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2015-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2015 IEEE International Conference on Consumer Electronics (ICCE)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICCE.2015.7066520","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 tries to address the problem of removing unwanted reflection layer from the image mixtures. These reflections may occur due to semi-reflective glass mediums. This demands separating the input image into the reflecting layer and the subject layer, also known as, background layer which is the actual scene itself. But decomposing a single input image into these two layers is a massively ill-posed problem with infinite combinations of decomposition. This type of problem falls under the category of blind source separation. There are ample classical separation approaches available in the literature which either requires multiple images or works on a single image with user assistance. In this paper we propose a method for separating the two layers from a single input image without any user or human intervention using some prior information about the reflective layer.