發(fā)布時(shí)間：2018-09-08 13:47

【摘要】：在近些年來,隨著遙感,GPS導(dǎo)航以及計(jì)算機(jī)等技術(shù)領(lǐng)域的發(fā)展,多技術(shù)集成的無(wú)人機(jī)遙感技術(shù)順勢(shì)而起。其應(yīng)用領(lǐng)域也不斷拓展,在動(dòng)態(tài)監(jiān)測(cè),數(shù)字化建設(shè)以及災(zāi)害應(yīng)急等多領(lǐng)域有著成熟的應(yīng)用。對(duì)于無(wú)人機(jī)這種尺度變化及旋轉(zhuǎn)角都大的影像,關(guān)于其拼接匹配技術(shù)的研究也比較多。但是,針對(duì)大比例尺無(wú)人機(jī)影像提取高分辨率DEM的研究相對(duì)較少,如何根據(jù)影像地形特征快速建立數(shù)字可視化模型是無(wú)人機(jī)遙感影像應(yīng)用成果的體現(xiàn),也是其深入各領(lǐng)域發(fā)展的前提。文章首先概述了當(dāng)前無(wú)人機(jī)影像在三維制作上的應(yīng)用研究現(xiàn)狀。針對(duì)影像分辨率高,且具有一定重疊度等特點(diǎn),提出了基于無(wú)人機(jī)影像快速建立三維數(shù)字化模型的方法。 研究工作包括以下幾個(gè)方面： (1)針對(duì)當(dāng)前大比例尺無(wú)人機(jī)影像制作DEM數(shù)據(jù)的方法研究中存在精度不高以及缺乏有效的質(zhì)量評(píng)定等問題,文章引入了中誤差及地面粗糙度等評(píng)價(jià)因子對(duì)利用地形特征點(diǎn)提取高分辨率DEM數(shù)據(jù)進(jìn)行了質(zhì)量評(píng)價(jià),從而避免了DEM數(shù)據(jù)在地形數(shù)字化應(yīng)用方面的盲目性。 (2)考慮無(wú)人機(jī)影像地形高程差的影響,從引起影像變形誤差的原理出發(fā),分析了無(wú)人機(jī)影像幾何變形的過程,并且主要對(duì)由外方位元素引起的誤差進(jìn)行了糾正處理。利用特征控制點(diǎn)并結(jié)合多項(xiàng)式及共線方程方法對(duì)不同地形加以處理,使得后期影像數(shù)據(jù)在三維應(yīng)用上的結(jié)果更加理想。 (3)針對(duì)大數(shù)據(jù)量,尤其是對(duì)于大規(guī)模城市區(qū)域建立數(shù)字化模型時(shí),建模速度慢,顯示效率不高。本文結(jié)合傳統(tǒng)分塊算法和三角網(wǎng)生成算法,提出了一種自適用閾值分塊構(gòu)網(wǎng)算法,從而提高了建模速度,在保證一定質(zhì)量的條件下也提高了自動(dòng)化處理水平。 (4)結(jié)合傳統(tǒng)二叉樹和四叉樹思想建立了一種基于觀察點(diǎn)的多分辨率模型,根據(jù)地形特征點(diǎn)與觀察點(diǎn)的距離大小,分析其分辨率的動(dòng)態(tài)變換關(guān)系,根據(jù)地形復(fù)雜度合適采用細(xì)化、簡(jiǎn)化處理,有效解決了三維可視化模型顯示效率不高的問題。應(yīng)用該模型分別在百度地圖與Google earth上實(shí)現(xiàn)了影像二維疊加以及三維航跡可視化模擬。圖48幅,表4個(gè),參考文獻(xiàn)52篇。
[Abstract]:In recent years, with the development of GPS navigation and computer technology, multi-technology integrated remote sensing technology of UAV has come into being. Its application field also expands unceasingly, has the mature application in the dynamic monitoring, the digitization construction and the disaster emergency and so on many fields. For UAV images with large scale variation and rotation angle, there are more researches on the matching techniques. However, there is relatively little research on high-resolution DEM extraction from large scale UAV images. How to quickly establish a digital visualization model based on the terrain features of UAV images is the embodiment of UAV remote sensing image application results. It is also the premise of its in-depth development in various fields. Firstly, the paper summarizes the application and research status of UAV image in three-dimensional production. Aiming at the characteristics of high resolution and certain overlap, a method of building 3D digital model based on UAV image is proposed. The research work includes the following aspects: (1) there are some problems such as low precision and lack of effective quality evaluation in the research of DEM data for large scale UAV images. In this paper, the evaluation factors such as middle error and surface roughness are introduced to evaluate the quality of high resolution DEM data extracted from terrain feature points. Therefore, the blindness of DEM data in terrain digitization is avoided. (2) considering the influence of terrain elevation difference of UAV image, the process of geometric deformation of UAV image is analyzed based on the principle of causing image deformation error. The error caused by the external azimuth element is corrected. By using characteristic control points, polynomial and collinear equation method to deal with different terrain, the result of 3D application of late image data is more ideal. (3) aiming at the large amount of data, Especially for large-scale urban areas, the modeling speed is slow and the display efficiency is not high. In this paper, combining with traditional block algorithm and triangular mesh generation algorithm, a self-adaptive threshold partition algorithm is proposed, which improves the speed of modeling. The automatic processing level is also improved under the condition of certain quality. (4) combined with the traditional binary tree and quadtree idea, a multi-resolution model based on observation point is established, according to the distance between terrain feature point and observation point, The dynamic transformation relation of its resolution is analyzed. According to the terrain complexity, thinning and simplification are adopted, which can effectively solve the problem that the display efficiency of 3D visualization model is not high. The model is used to simulate the two dimensional superposition of image and the visualization of 3D track on Baidu map and Google earth respectively. 48 figures, 4 tables, 52 references.
【學(xué)位授予單位】：中南大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2013
【分類號(hào)】：P208;P231

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 谷寧;馮仲科;羅旭;;基于無(wú)人飛行器航測(cè)建立數(shù)字高程模型研究[J];北京林業(yè)大學(xué)學(xué)報(bào);2007年S2期

2 李軍;李永樹;蔡國(guó)林;;利用無(wú)人機(jī)影像制作地震災(zāi)區(qū)三維景觀圖[J];測(cè)繪工程;2012年01期

3 崔紅霞,林宗堅(jiān),孫杰;大重疊度無(wú)人機(jī)遙感影像的三維建模方法研究[J];測(cè)繪科學(xué);2005年02期

4 范承嘯;韓俊;熊志軍;趙毅;;無(wú)人機(jī)遙感技術(shù)現(xiàn)狀與應(yīng)用[J];測(cè)繪科學(xué);2009年05期

5 陳飛;吳英男;;基于數(shù)字高程模型和遙感影像的三維可視化[J];測(cè)繪科學(xué);2009年S2期

6 胡榮明;張敏;劉潘;;高分辨率無(wú)人機(jī)低空影像DEM的建立及其精度研究[J];測(cè)繪科學(xué);2011年04期

7 張麗麗;王小平;張瑛;;基于無(wú)人機(jī)影像生產(chǎn)高精度DEM的實(shí)踐[J];測(cè)繪技術(shù)裝備;2009年01期

8 崔紅霞,林宗堅(jiān),孫杰;無(wú)人機(jī)遙感監(jiān)測(cè)系統(tǒng)研究[J];測(cè)繪通報(bào);2005年05期

9 劉異;李玉霞;童玲;;無(wú)地面控制點(diǎn)的無(wú)人機(jī)遙感影像幾何校正算法[J];測(cè)繪通報(bào);2012年07期

10 卜彥龍;潘亮;彭輝;沈林成;;未知DEM下的機(jī)載SAR圖像幾何校正方法研究[J];測(cè)繪學(xué)報(bào);2009年01期

，

本文編號(hào)：2230699