本文關(guān)鍵詞： 沖溝侵蝕 資源三號衛(wèi)星 影像融合 沖溝提取　出處：《魯東大學》2016年碩士論文　論文類型：學位論文

【摘要】：近年來,隨著空間信息及測量技術(shù)的發(fā)展,特別是高分遙感、立體攝影測量、雷達干涉測量、RTK-GPS測量、激光三維掃描等新技術(shù)的涌現(xiàn),為沖溝侵蝕的研究提供了大量的多源數(shù)據(jù),豐富了沖溝侵蝕研究的方法和手段。但這些數(shù)據(jù)多單獨應(yīng)用于沖溝侵蝕研究,各種數(shù)據(jù)的結(jié)合應(yīng)用只存在于獨立的交叉驗證,缺乏真正的融合。緊密融合多源數(shù)據(jù),構(gòu)建基于多源數(shù)據(jù)融合的沖溝參數(shù)提取方法成為目前迫切需要解決的問題�；诖�,本文選擇沖溝侵蝕廣泛分布的庵里水庫西緣流域為研究區(qū)域,探討多源遙感數(shù)據(jù)的融合技術(shù),建立基于多源數(shù)據(jù)融合的提取沖溝參數(shù)方法,以期拓寬沖溝侵蝕研究方法,提高沖溝研究精度。獲得了以下結(jié)論:(1)比較三種全色與多光譜影像融合方法,HSV變換影像圖像亮度過高、光譜損失嚴重,目視效果較差。PCA變換影像信息量較豐富,但清晰度最差,光譜損失程度較大,目視效果最差。GS光譜銳化影像目視效果最好,光譜保持能力好,信息量豐富及清晰度較高。GS光譜銳化融合方法是一種適宜資源三號衛(wèi)星全色與多光譜影像融合的方法。(2)基于GS光譜銳化、HSV變換及PCA變換影像提取的沖溝數(shù)量均多于原始影像提取的沖溝數(shù)量,比基于原始多光譜影像解譯的沖溝更準確。結(jié)合實測沖溝參數(shù)對基于融合影像提取的沖溝參數(shù)進行精度檢驗時發(fā)現(xiàn),基于GS光譜銳化圖像提取的沖溝參數(shù)精度最高,GS光譜銳化融合方法是一種適合沖溝參數(shù)提取的融合算法。(3)采用SRM乘法融合法和Gram-Schmidt光譜銳化方法、HSV變換三種方法對資源三號衛(wèi)星多光譜圖像進行DEM與多光譜影像融合,三種融合方法都將SRM數(shù)據(jù)綜合得到原始圖像中,實現(xiàn)了圖像的正立體化,提高了遙感圖像解譯地形參數(shù)的能力;SRM乘法校正后圖像光譜保持能力好,適宜定量遙感分析,并且對平坦地區(qū)地物光譜信息保持程度最好;HSV融合圖像光譜失真明顯,但信息豐富、清晰度好且地形正立體感強,適合地貌學研究;GS融合圖像正立體感明顯,光譜失真較明顯,信息量與清晰度不如HSV融合法,適合通用制圖應(yīng)用。(4)選取HSV變換方法對DEM數(shù)據(jù)及多光譜數(shù)據(jù)進行了融合,融合結(jié)果一方面克服了原始圖像表現(xiàn)出的地形視覺混淆,另一方面,賦予了圖像豐富的三維地形信息,有利于對沖溝的判讀;基于融合影像,以結(jié)合溝沿線、溝底線坡度閾值與目視解譯的沖溝提取方式,提取實驗區(qū)76條沖溝,沖溝總面積1.13km2,溝壑密度5.2km/km2,為極強烈侵蝕;以基于DOM影像提取的沖溝作為標準沖溝,對基于融合影像提取的沖溝參數(shù)進行精度檢驗,基于融合影像提取的沖溝有效溝沿線及溝底線比率均為91%,明顯高于原始影像,基于融合影像提取的沖溝參數(shù)精度高于多光譜影像;多光譜影像與地形數(shù)據(jù)融合方法可以提高沖溝參數(shù)解譯精度。
[Abstract]:In recent years, with the development of spatial information and measurement technology, especially high-score remote sensing, stereo photogrammetry, radar interferometry RTK-GPS measurement, laser three-dimensional scanning and other new technologies, a large number of multi-source data have been provided for the study of gully erosion. It enriches the methods and means of gully erosion research, but most of these data are used separately in gully erosion research, and the combination of all kinds of data only exists in independent cross-validation, lacking of true fusion. It has become an urgent problem to construct a method of extracting gully parameters based on multi-source data fusion. Based on this, this paper selects the western edge basin of Anli Reservoir, where gully erosion is widely distributed, as the research area. This paper discusses the fusion technology of multi-source remote sensing data and establishes a method of extracting gully parameters based on multi-source data fusion in order to broaden the research method of gully erosion. In order to improve the precision of gully research, the following conclusions are obtained: (1) comparing the three fusion methods of panchromatic and multispectral images, the HSV transform images are too bright, the spectral loss is serious, the visual effect is poor. The PCA transform image is rich in information, but the definition is the worst. The degree of spectral loss is large, the visual effect is the worst. GS spectral sharpening image has the best visual effect, and the spectral retention ability is good. The method of spectral sharpening fusion with rich information and high definition. GS is a suitable method for fusion of panchromatic and multispectral images of satellite No. 3.) based on GS spectral sharpening and PCA transform, the number of gullies extracted is many. The number of gullies extracted from the original image, It is more accurate than that based on the interpretation of the original multispectral image. GS spectral sharpening fusion method based on GS spectral sharpening image extraction is a fusion algorithm suitable for gully parameter extraction. It adopts SRM multiplication fusion method and Gram-Schmidt spectral sharpening method. The multispectral image of Resource-3 satellite is fused with DEM and multispectral image. All of the three fusion methods synthesize the SRM data into the original image, realize the orthotropic of the image, improve the ability of interpreting the terrain parameters of the remote sensing image and improve the spectral retention ability of the image after the correction by the SRM multiplication, which is suitable for quantitative remote sensing analysis. The spectral distortion of HSV fusion image is obvious, but the information is rich, the definition is good and the terrain is positive stereoscopic, which is suitable for geomorphology study of GS fusion image, and the spectral distortion is obvious. The amount and clarity of information is inferior to that of HSV fusion method, which is suitable for general cartographic application. (4) HSV transform method is selected to fuse DEM data and multispectral data. The fusion results not only overcome the terrain visual confusion of the original image, but also overcome the visual confusion of the original image on the other hand. The images are endowed with abundant 3D terrain information, which is beneficial to the interpretation of the gully, and based on the fusion image, 76 gullies are extracted from the experimental area by combining the gully baseline slope threshold with visual interpretation. The total area of the gully is 1.13km2, and the density of the gully is 5.2km2, which is very strong erosion. The gully extracted based on the DOM image is taken as the standard gully, and the precision of the gully parameters extracted from the fusion image is tested. The ratio of effective gully and furrow bottom line based on fusion image is 91R, which is obviously higher than that of original image, and the precision of gully parameter extraction based on fusion image is higher than that of multispectral image. Multispectral image and terrain data fusion method can improve the precision of gully parameter interpretation.
【學位授予單位】：魯東大學
【學位級別】：碩士
【學位授予年份】：2016
【分類號】：S157.1

【參考文獻】

相關(guān)期刊論文 前10條

1 何福紅;王濤;范擎宇;張振華;吳孟泉;;地形和遙感圖像融合技術(shù)在沖溝參數(shù)提取中的應(yīng)用——以棲霞市庵里水庫東緣流域為例[J];地理研究;2015年11期

2 姜濤;朱文泉;周夏飛;鄭周濤;張東海;馬國霞;趙學濤;;一種融合地形陰影模型和YUV亮度分量的遙感圖像反立體校正方法[J];北京師范大學學報(自然科學版);2015年S1期

3 朱昕宇;孫明偉;王樹根;余磊;郭清;;衛(wèi)星影像反立體改正算法比較探究[J];華中師范大學學報(自然科學版);2015年05期

4 李佳佳;熊東紅;盧曉寧;董一帆;蘇正安;翟娟;楊丹;;基于RTK-GPS技術(shù)的干熱河谷沖溝溝頭形態(tài)特征[J];山地學報;2014年06期

5 吳曉萍;楊武年;李國明;;資源三號衛(wèi)星正視全色與多光譜影像融合及評價[J];物探化探計算技術(shù);2014年01期

6 何福紅;高丙艦;王煥芝;王瑞;賽莉莉;;基于GIS的侵蝕沖溝與地貌因子的關(guān)系[J];地理研究;2013年10期

7 宋萍;齊偉;徐柏琪;李哲;路超;張圣武;;膠東山區(qū)景觀格局與環(huán)境因子關(guān)系研究——以山東省棲霞市為例[J];中國生態(tài)農(nóng)業(yè)學報;2013年03期

8 高丙艦;何福紅;王瑞;唐彥君;;基于高分辨率影像的沖溝發(fā)育及其時空變化研究[J];地理空間信息;2013年01期

9 李德仁;;我國第一顆民用三線陣立體測圖衛(wèi)星——資源三號測繪衛(wèi)星[J];測繪學報;2012年03期

10 胡龍華;徐豁;;城市衛(wèi)星遙感圖像融合處理質(zhì)量評價研究[J];華北科技學院學報;2012年01期

相關(guān)博士學位論文 前1條

1 李偉;像素級圖像融合方法及應(yīng)用研究[D];華南理工大學;2006年

相關(guān)碩士學位論文 前4條

1 徐莉莉;基于天繪衛(wèi)星的遙感影像融合研究[D];長安大學;2014年

2 李琳琳;基于GIS的典型黑土區(qū)溝蝕動態(tài)及其影響因子分析[D];吉林大學;2010年

3 肖晨超;基于DEM的黃土地貌溝沿線特征研究[D];南京師范大學;2007年

4 劉宇瓊;基于像素級多源遙感影像融合技術(shù)的研究[D];長安大學;2007年

，

本文編號：1520182