{"title":"Underwater image enhancement using single scale retinex on a reconfigurable hardware","authors":"Alex Raj S. M, M. Supriya","doi":"10.1109/SYMPOL.2015.7581166","DOIUrl":null,"url":null,"abstract":"In this paper, hardware implementation of an improved image enhancement technique using single scale retinex algorithm is proposed. Considering the real scene, the dynamic range of digital camera is narrower and hence contrast correction is required to reproduce the information in darker regions. In the proposed method, an input RGB color image is converted to YCbCr color space, and Y and Cr component is modified as the variations in blue components are nominal. In enhancing Y and Cr component, the Gaussian surround function is convolved to it and then the difference between scaled version of Y and Cr component and the convolved one is added to the original one. The algorithm is implemented in FPGA board. FPGA platform is preferred as it's ability to perform parallel algorithm due to it's inherent parallelism.","PeriodicalId":127848,"journal":{"name":"2015 International Symposium on Ocean Electronics (SYMPOL)","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2015-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"14","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2015 International Symposium on Ocean Electronics (SYMPOL)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/SYMPOL.2015.7581166","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 14
Abstract
In this paper, hardware implementation of an improved image enhancement technique using single scale retinex algorithm is proposed. Considering the real scene, the dynamic range of digital camera is narrower and hence contrast correction is required to reproduce the information in darker regions. In the proposed method, an input RGB color image is converted to YCbCr color space, and Y and Cr component is modified as the variations in blue components are nominal. In enhancing Y and Cr component, the Gaussian surround function is convolved to it and then the difference between scaled version of Y and Cr component and the convolved one is added to the original one. The algorithm is implemented in FPGA board. FPGA platform is preferred as it's ability to perform parallel algorithm due to it's inherent parallelism.