本文選題：GF- + 水體信息提取　； 參考：《國土資源遙感》2017年01期

【摘要】：大尺度高精度山區(qū)河流信息提取是我國干旱區(qū)水資源開發(fā)利用的關(guān)鍵技術(shù),而利用遙感影像提取水資源信息存在水體與山區(qū)陰影難以區(qū)分的瓶頸。以GF-1號衛(wèi)星2 m分辨率全色波段影像和8 m分辨率多光譜影像為數(shù)據(jù)源,選取新疆特克斯河流域巴喀勒克水庫為研究區(qū),提出改進(jìn)的陰影水體指數(shù)法(modified shade water index,MSWI)進(jìn)行水體信息提取;同時運(yùn)用單波段閾值法、NDWI法、單波段法與陰影水體指數(shù)法(shade water indes,SWI)相結(jié)合的決策樹分類法(簡稱SWI)以及單波段法與MSWI相結(jié)合的決策樹分類法(簡稱MSWI)分別對研究區(qū)水體信息進(jìn)行提取,并進(jìn)行了對比分析。研究結(jié)果表明,前2種方法與SWI和MSWI法相比,效果稍差;而SWI和MSWI法分類效果較好,其中MSWI比SWI法分類總精度高0.94%,提高了高分辨率遙感影像的解譯精度,可為國產(chǎn)高分系列衛(wèi)星影像在干旱區(qū)水資源信息提取中的應(yīng)用提供技術(shù)支持。
[Abstract]:The large scale and high precision mountain river information extraction is the key technology of water resources development and utilization in the arid area of our country. While using remote sensing image to extract water resources information is difficult to distinguish between water body and mountain shadow. Taking GF-1 2 m resolution panchromatic image and 8 m resolution multi spectral image as data source, Xinjiang Turks are selected. The bakklek reservoir in the river domain is the research area, and the improved shadow water index method (modified shade water index, MSWI) is used to extract the water information. At the same time, the single band threshold method, NDWI method, the single band method and the shadow water index method (shade water indes, SWI) are combined with the decision tree classification method (SWI) and the single band method and MSWI. The decision tree classification method (MSWI) is used to extract the water information of the study area, and the results are compared. The results show that the first 2 methods are less effective than the SWI and MSWI methods, while the SWI and MSWI methods have better classification effect, and the total accuracy of MSWI is higher than that of the SWI method, and the interpretation of high resolution remote sensing images is improved. The accuracy can provide technical support for the application of domestic high score satellite images in the extraction of water resources information in arid areas.
【作者單位】： 新疆大學(xué)資源與環(huán)境科學(xué)學(xué)院綠洲生態(tài)教育部重點(diǎn)實(shí)驗(yàn)室;新疆交通職業(yè)技術(shù)學(xué)院;
【基金】：國防科技工業(yè)局高分辨率對地觀測重大專項(民用部分)項目“中亞地區(qū)跨境河流水資源利用開發(fā)遙感監(jiān)測系統(tǒng)”(編號:95-Y40B02-9001-13/15-03-01) 教育部新世紀(jì)優(yōu)秀人才支持計劃項目“區(qū)域水鹽遙感監(jiān)測與模擬方法研究”(編號:NCET-12-1075) 2014年新疆研究生科研創(chuàng)新項目“基于國產(chǎn)高分衛(wèi)星影像的水資源開發(fā)利用遙感監(jiān)測系統(tǒng)”(編號:XJGRI2014022)共同資助
【分類號】：TP751

【相似文獻(xiàn)】

相關(guān)期刊論文 前4條

1 潘琛;杜培軍;張海榮;;決策樹分類法及其在遙感圖像處理中的應(yīng)用[J];測繪科學(xué);2008年01期

2 李爽,丁圣彥,錢樂祥;決策樹分類法及其在土地覆蓋分類中的應(yīng)用[J];遙感技術(shù)與應(yīng)用;2002年01期

3 演克武;張磊;孫強(qiáng);;決策樹分類法中ID3算法在航空市場客戶價值細(xì)分中的應(yīng)用[J];商業(yè)研究;2008年03期

4 ;[J];;年期

相關(guān)碩士學(xué)位論文 前1條

1 孫明;基于遙感技術(shù)的雅魯藏布江源區(qū)植被類型及覆蓋度研究[D];南京信息工程大學(xué);2011年

，

本文編號：1950941