K. S. Himaja Chowdary, M. Kalaiyarasi, Swaminathan Saravanan
{"title":"Gated Recurrent Unit RNN based Non-negative Tucker Decomposition for Satellite Image Compression","authors":"K. S. Himaja Chowdary, M. Kalaiyarasi, Swaminathan Saravanan","doi":"10.1109/TEECCON54414.2022.9854846","DOIUrl":null,"url":null,"abstract":"Satellite images are often volumetric, requiring a lot of storage and transmission space and time. In this paper, a Gated Recurrent Unit RNN based NTD method has been proposed for satellite image compression. RNN is used to convert spectral sensor into small scale spectral sensor. Entropy encoding is performed for final compression. The proposed method is compared to the standard NTD in the wavelet domain, the computing efficiency is improved by 56.40% while compromising just -0.58 dB of PSNR.","PeriodicalId":251455,"journal":{"name":"2022 Trends in Electrical, Electronics, Computer Engineering Conference (TEECCON)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-05-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2022 Trends in Electrical, Electronics, Computer Engineering Conference (TEECCON)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/TEECCON54414.2022.9854846","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 2
Abstract
Satellite images are often volumetric, requiring a lot of storage and transmission space and time. In this paper, a Gated Recurrent Unit RNN based NTD method has been proposed for satellite image compression. RNN is used to convert spectral sensor into small scale spectral sensor. Entropy encoding is performed for final compression. The proposed method is compared to the standard NTD in the wavelet domain, the computing efficiency is improved by 56.40% while compromising just -0.58 dB of PSNR.