本文選題：暗通道 + 高程信息提取��； 參考：《南京理工大學(xué)》2017年碩士論文

【摘要】：我國是一個(gè)自然災(zāi)害頻發(fā)的國家,其中,滑坡、泥石流就是兩種典型的地質(zhì)災(zāi)害,它們危害最大、分布最廣、發(fā)生最頻繁。因此,對(duì)滑坡泥石流的檢測(cè)很有必要,本文提出了一種基于暗通道原理的單幅遙感圖像高程值提取算法,同時(shí)給出了一種克服山脈陰影影響的解決策略,得到的高程信息可以應(yīng)用于滑坡、泥石流自然災(zāi)害的檢測(cè)。從下面四個(gè)方面闡述本文的主要研究工作和研究成果:(1)通過對(duì)大量山脈遙感圖像的分析,驗(yàn)證了暗通道先驗(yàn)原理同樣適用于山脈遙感圖像,并將暗通道原理結(jié)合單幅山脈遙感圖像來提取山脈高程。(2)利用了山脈的地理信息來分析陰影分布情況和陰影特點(diǎn),闡述了幾種普遍的陰影檢測(cè)算法,引入機(jī)器學(xué)習(xí)檢測(cè)陰影的思想,并對(duì)陰影檢測(cè)的結(jié)果進(jìn)行定量和定性分析,選擇最佳的陰影檢測(cè)算法。(3)將輸出的高程信息用于滑坡泥石流檢測(cè),對(duì)疑似滑坡泥石流的區(qū)域,結(jié)合輸出的高程信息圖,分析區(qū)域周圍的高程信息,判斷是否發(fā)生滑坡泥石流。(4)開發(fā)實(shí)現(xiàn)了一個(gè)全新的單幅山脈遙感圖像提取高程及應(yīng)用系統(tǒng),融合了基于暗通道原理的高程提取算法,基于陰影調(diào)整的高程信息提取算法,以及結(jié)合高程信息的滑坡泥石流檢測(cè),為將來的研究提供簡(jiǎn)便有力的工具。對(duì)大量山脈遙感圖像進(jìn)行了定量和定性分析,實(shí)驗(yàn)結(jié)果表明,本文提出的算法能夠有效提取出包含陰影的單幅山脈遙感圖像的相對(duì)高程信息,并且該高程信息能夠較好地用于滑坡泥石流自然災(zāi)害的檢測(cè)。
[Abstract]:China is a country with frequent natural disasters, among which landslide and debris flow are two typical geological disasters, which are the most harmful, most widely distributed and most frequent. Therefore, it is necessary to detect landslide and debris flow. In this paper, a single remote sensing image elevation extraction algorithm based on dark channel principle is proposed, and a solution strategy to overcome the influence of mountain shadow is given. The obtained elevation information can be applied to the detection of landslide and debris flow natural disasters. By analyzing a large number of remote sensing images of mountain ranges, it is proved that the prior principle of dark channels is also applicable to remote sensing images of mountain ranges. The principle of dark channel is combined with a single remote sensing image of mountain range to extract mountain elevation. (2) using the geographic information of mountain range to analyze the shadow distribution and shadow characteristics, several common shadow detection algorithms are described. The idea of detecting shadow by machine learning is introduced, and the results of shadow detection are analyzed quantitatively and qualitatively, and the best shadow detection algorithm. Combined with the output elevation information map, the height information around the area is analyzed, and the landslide and debris flow is judged. 4) A completely new remote sensing image of single mountain range is developed to extract elevation and its application system. The algorithm of height extraction based on dark channel principle, height information extraction algorithm based on shadow adjustment, and landslide and debris flow detection combined with elevation information are combined to provide a simple and powerful tool for future research. Quantitative and qualitative analysis of a large number of remote sensing images of mountain ranges has been carried out. The experimental results show that the proposed algorithm can effectively extract the relative elevation information of a single mountain remote sensing image with shadow. The elevation information can be used to detect natural disasters of landslide and debris flow.
【學(xué)位授予單位】：南京理工大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP751

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 曾敬文;王學(xué)文;侯文廣;劉光;;連續(xù)局部閾值分割的陰影檢測(cè)方法[J];測(cè)繪科學(xué);2016年11期

2 徐旭;張風(fēng)麗;王國軍;符喜優(yōu);沙敏敏;李志坤;邵蕓;陳龍永;梁興東;;基于高亮特征匹配的雙視向SAR圖像建筑物高度提取[J];遙感技術(shù)與應(yīng)用;2016年01期

3 錢瑤;唐立娜;趙景柱;;基于遙感的建筑物高度快速提取研究綜述[J];生態(tài)學(xué)報(bào);2015年12期

4 胡甚;張基宏;梁永生;柳偉;陳小平;;一種基于變分權(quán)值優(yōu)化的多尺度圖像融合方法[J];濟(jì)南大學(xué)學(xué)報(bào)(自然科學(xué)版);2016年01期

5 田建東;王占鵬;唐延?xùn)|;;靜態(tài)陰影檢測(cè)的研究進(jìn)展[J];信息與控制;2015年02期

6 石義方;汪小欽;孫振海;陳蕓芝;付乾坤;;基于陰影的資源三號(hào)衛(wèi)星數(shù)據(jù)城市建筑物高度估算[J];地球信息科學(xué)學(xué)報(bào);2015年02期

7 周雨薇;楊平呂;陳強(qiáng);孫權(quán)森;;基于MTF和變分的全色與多光譜圖像融合模型[J];自動(dòng)化學(xué)報(bào);2015年02期

8 方菊芹;陳帆;和紅杰;尹忠科;;結(jié)合局部分類水平集與顏色特征的遙感影像陰影檢測(cè)[J];自動(dòng)化學(xué)報(bào);2014年06期

9 AmirReza SHAHTAHMASSEBI;YANG Ning;WANG Ke;Nathan MOORE;SHEN Zhangquan;;Review of Shadow Detection and De-shadowing Methods in Remote Sensing[J];Chinese Geographical Science;2013年04期

10 王媛媛;陳旺;張茂軍;王煒;徐瑋;;折反射全向圖像與遙感圖像配準(zhǔn)的建筑物高度提取算法[J];計(jì)算機(jī)應(yīng)用;2011年09期

，

本文編號(hào)：1853415