预览加载中,请您耐心等待几秒...
1/2
2/2

在线预览结束,喜欢就下载吧,查找使用更方便

如果您无法下载资料,请参考说明:

1、部分资料下载需要金币,请确保您的账户上有足够的金币

2、已购买过的文档,再次下载不重复扣费

3、资料包下载后请先用软件解压,在使用对应软件打开

基于分布式信源编码和感兴趣区域编码的干涉多光谱图像压缩(英文) Interferometrichyperspectralimagingisapowerfultoolforvariousremotesensingandimagingapplicationswithhighspatialandspectralresolutions.However,themassiveamountsofdatageneratedbythesesystemshavebeenasignificantchallengeforstorage,transmissionandprocessingofthedata.Inthispaper,weproposeaninterferencehyperspectralimagecompressionmethodbasedondistributedsourcecodingandregionofinterestcoding. Firstly,weutilizedistributedsourcecodingtoexploitthestatisticaldependenciesbetweenneighboringpixels.Inthismethod,thedataateachpixelisindependentlyquantizedandtransmitted.Thecomplementaryinformationisusedatthedecodertoreconstructtheoriginalhyperspectralimage.Thisapproachhastheadvantageoflowcomplexityencodinganddecoding,lowencodingrate,aswellasgoodcompressionperformance.Sincethecompressionratioobtainedbydistributedsourcecodingislowerthanwhatisrequired,especiallyforhyperspectralimages,theproposedmethodusesmultipledistributedencodingschemestoachieveahighercompressionratio. Secondly,weapplyregionofinterestcoding(ROI)tofurtherimprovecompressionperformance.TheROIreferstoaspecificareawithinthehyperspectralimagethatcontainsimportantinformation,suchastargetobjects.Intheproposedmethod,wecombinethedistributedcodingmethodandROIcodingtoprovideamoretargetedcompressionforspecificregionsintheimage.WeassumethattheROIisidentifiedbytheuserbeforehandandistransmittedtotheencoder.TheencoderthencompressestheROIseparatelytoahighercompressionratio.Incontrast,thenon-ROIiscompressedusingdistributedsourcecoding.Atthedecoder,thedecoderreconstructstheROIandcombinesitwiththereconstructednon-ROItoobtainthefinalreconstructedhyperspectralimage.TheproposedmethodfurtherimprovescompressionperformancebyexploitingspatialredundancybetweenpixelsoftheROI. Wesimulateourproposedmethodonthreehyperspectralimagesandcompareourperformancewithseveralstate-of-the-artmethods.Experimentalresultsshowthattheproposedmethodachievesahighercompressionratiocomparedtothecomparedmethodswhilemaintainingasimilarorbet