《基于特征区域统计的多光谱遥感影像海岛瞬时岸线提取（英文版）》.pdfVIP

Vol. 16 No.1 MarineScienceBulletin May 2014 Islandinstantaneouscoastlineextraction basedon thecharacteristicsofregionalstatisticsof multispectralremotesensingimage 1,2 2 2 2 2 2 WANG Fen ,LIUShu-ming ,LUWen-hu ,DUQiong-wei ,JIANGWei-nan ,LIUJin 1. College of Oceanography and Environmental Science of Xiamen University,Xiamen 361005, Fujian province, China; 2.NotionalMarineDateandInformationService,Tianjing300171, China Abstract: This study adopted IKONOS remote sensing images and selected spectralcharacteristicareas,throug regional pixelstatisticsandcalculating weight coefficients of eac band, processed the images wit the spectral normalized method, whic made the features of islands, land and water features moreobviouslyinthe images.Onthis basis,theOTUSwas usedto determine the optimal segmentation threshold, andthe normalization image binarizationwas made,thusthe islandcoastlinewas extracted.This method used the characteristic curve method to separate the land and water, obtainedthebinarizationimagesandmaintainedtheoriginaledgeeffectively. The coastline that was extracted by Binary Morphology was continuous, reliable and hig signal-to-noise ratio.The results showed that this method couldextractthecoastlinefast,simplyandeffectively,whic hadthepractical value. Keywords: coastline extraction; islands; spectral characteristics of regional statistics; BinaryMorphology;IKONOS Introduction China is ric of islands, therefore protection and management of the island has become increasingly important. Island coastline, a basic feature of the isl