【摘要】：【目的】利用不同分類精度的指數(shù)模型分析南京2001—2015年的城市化進(jìn)程。【方法】借助2001年Landsat7 ETM+,2006、2010年Landsat5 TM及2015年Landsat8 OLI南京地區(qū)遙感影像數(shù)據(jù),首先通過歸一化差異水體指數(shù)(the modified normalized difference water index,MNDWI)提取水體信息,隨后利用城市指數(shù)(the urban index,UI)、歸一化建筑指數(shù)(the normalized difference built-up index,NDBI)、建筑指數(shù)(the index-based built-up index,IBI)、基于紅光的建筑指數(shù)(the visible red-based built-up index,Vr NIR-BI)、基于綠光的建筑指數(shù)(the visible green-based built-up index,Vg NIR-BI)模型提取不透水層,結(jié)合最大類間差分法將影像分為建成區(qū)和非建成區(qū)兩類,最后融合分類結(jié)果形成南京地區(qū)城市化專題地圖。【結(jié)果】從驗(yàn)證結(jié)果來看,所有模型的總體精度均高于64.33%,其中UI、NDBI模型具有總體最高精度。UI模型精度最低69.67%(2001年),最高85%(2010年);NDBI模型精度最低72.00%(2001年),最高82.60%(2010年)。在UI模型和NDBI模型下城市化發(fā)展水平,分別從2001年的11.55%和17.66%攀升至2015年的20.50%和25.60%�！窘Y(jié)論】在精度最高的UI指數(shù)和NDBI指數(shù)模型下,南京市除高淳區(qū)和溧水區(qū)之外的各個行政區(qū)的建成區(qū)增量前4名分別是江寧區(qū)、六合區(qū)、浦口區(qū)和棲霞區(qū)。
[Abstract]:[objective] to analyze the urbanization process of Nanjing from 2001 to 2015 by using the index model with different classification accuracy. [methods] the remote sensing images of Nanjing in 2001, Landsat5 TM in 2010 and Landsat8 OLI in 2015 were used. Firstly, the water information is extracted by normalized differential water body index (the modified normalized difference water). Then the impervious layer was extracted by using the city index (the urban index UI, the normalized building index (the normalized difference built-up index NDBI, the index-based built-up index IBI, the red light based (the visible red-based built-up index Vr NIR-BI and the green light based building index (the visible green-based built-up indexVg NIR-BI. Combined with the maximum inter-class difference method, the image is divided into two categories: the established area and the non-built area. Finally, the fusion and classification results are used to form the thematic map of urbanization in Nanjing. [results] from the verification results, The overall accuracy of all models is higher than that of 64.33. The UINDBI model has the highest overall accuracy. The UI model has the lowest precision of 69.67% (2001), the highest accuracy of 85% (2010) and the lowest accuracy of 72.00% (2001) and 82.60% (2010) of the NDBI model. The level of urbanization in UI model and NDBI model rose from 11.55% and 17.66% in 2001 to 20.50% and 25.60% in 2015, respectively. [conclusion] under the most accurate UI index and NDBI index model, In addition to Gaochun District and Lishui District in Nanjing, the top four administrative districts are Jiangning District, Liuhe District, Pukou District and Qixia District respectively.
【作者單位】： 南京林業(yè)大學(xué)林學(xué)院南方現(xiàn)代林業(yè)協(xié)同創(chuàng)新中心;
【基金】：國家林業(yè)局“948”項(xiàng)目(2014-04-25) 江蘇高校品牌專業(yè)建設(shè)工程資助項(xiàng)目(PPZY2015A062) 國家自然科學(xué)基金項(xiàng)目(31270587,31670552) 江蘇高校優(yōu)勢學(xué)科建設(shè)工程資助項(xiàng)目(PAPD) 江蘇省高校研究生科研創(chuàng)新計劃項(xiàng)目(KYLX15_0908)
【分類號】：P237

【相似文獻(xiàn)】

相關(guān)碩士學(xué)位論文 前1條

1 姜海蘭;利用多時相Landsat衛(wèi)星影像直接提取沙漠化區(qū)域的簡便方法[D];蘭州大學(xué);2015年

，

本文編號：2169692