本文選題：地面三維激光掃描技術(shù) + 點云濾波　； 參考：《長安大學(xué)》2017年碩士論文

【摘要】：近年來,地面三維激光掃描(Terrestrial Laser Scanning,TLS)技術(shù)以其快速、高密度、高精度等特點被廣泛應(yīng)用于變形監(jiān)測領(lǐng)域。這項技術(shù)在極大地方便了測繪外業(yè)的同時,將主要工作放在了內(nèi)業(yè)數(shù)據(jù)處理上,其關(guān)鍵步驟有點云拼接、點云濾波、建立模型、多期模型配準(zhǔn)和形變量解算,其中,點云濾波和多期模型配準(zhǔn)問題一直是研究的難點和熱點。本文對三維激光掃描技術(shù)的特點及分類、地面三維激光掃描技術(shù)的原理、數(shù)據(jù)處理和應(yīng)用做了系統(tǒng)介紹,并在詳細(xì)介紹了地面三維激光掃描數(shù)據(jù)濾波和DEM(Digital Elevation Model)模型配準(zhǔn)現(xiàn)狀的基礎(chǔ)上,對點云數(shù)據(jù)濾波和無控制DEM配準(zhǔn)進行了研究。首先,在二維聚類算法的基礎(chǔ)上進行研究改進,提出了適用于點云數(shù)據(jù)處理的三維點云聚類濾波算法,有效避免了二維聚類算法使用中進行點云投影時不可避免的點云空間結(jié)構(gòu)信息的損失,充分利用了TLS點云數(shù)據(jù)的高維度和高密度特點,并通過實驗,將其與點云濾波處理常用的曲率平滑濾波算法、移動最小二乘趨勢面法進行比較,驗證了該算法在處理地形復(fù)雜的茂密植被區(qū)點云數(shù)據(jù)的優(yōu)勢;其次,針對無控制DEM配準(zhǔn)問題,在現(xiàn)有迭代最近點(Iterations closest point,ICP)算法、最小高差(Least Z-Difference,LZD)算法和最小二乘3D表面匹配(Least squares 3D surface matching,LS3D)算法的基礎(chǔ)上,用不同地形特征的仿真數(shù)據(jù)對進行實驗,比較這三種算法在配準(zhǔn)精度、配準(zhǔn)效率和拉入范圍方面的配準(zhǔn)性能,為多時相無控制DEM數(shù)據(jù)的配準(zhǔn)提供參考。然后使用上述方法對雞冠嶺地面三維激光掃描數(shù)據(jù)進行處理,解算其在10個月內(nèi)的形變量并進行了相應(yīng)的分析。
[Abstract]:In recent years, Terrestrial Laser scanning TLSs (TLSs) technology has been widely used in deformation monitoring due to its rapid, high density and high accuracy. This technology greatly facilitates the field of surveying and mapping at the same time, the main work is on the internal data processing, its key steps are cloud splicing, point cloud filtering, modeling, multi-phase model registration and shape variable calculation, in which, Point cloud filtering and multi-phase model registration are always the difficulties and hot spots. In this paper, the characteristics and classification of 3D laser scanning technology, the principle, data processing and application of 3D laser scanning technology are systematically introduced. Based on the detailed introduction of 3D laser scanning data filtering and digital elevation model registration, the point cloud data filtering and uncontrolled Dem registration are studied. Firstly, based on the research and improvement of two-dimensional clustering algorithm, a 3D point cloud clustering filtering algorithm is proposed, which is suitable for point cloud data processing. The loss of point cloud spatial structure information which is unavoidable in the point cloud projection in the use of two-dimensional clustering algorithm is effectively avoided, and the characteristics of high dimension and high density of TLS point cloud data are fully utilized, and the experiments are carried out. Compared with point cloud filtering and moving least square trend surface method, this algorithm has the advantages in dealing with point cloud data in dense vegetation areas with complex terrain. Secondly, aiming at the problem of uncontrolled Dem registration, it is proved that this algorithm has advantages in dealing with point cloud data in dense vegetation areas with complex terrain. On the basis of the existing iterations closest points algorithm, the least elevation difference algorithm and the least square 3D surface matching algorithm, the simulation data of different terrain features are used to compare the registration accuracy of the three algorithms. The registration efficiency and pull-in range provide a reference for the registration of multi-phase uncontrolled Dem data. Then, the above method is used to process the 3D laser scanning data of Jiguanling ground, and the shape variables within 10 months are calculated and the corresponding analysis is carried out.
【學(xué)位授予單位】：長安大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：P225.2

【參考文獻】

相關(guān)期刊論文 前10條

1 曾繁軒;李亮;;基于Lagrange算子與曲面擬合的點云濾波研究[J];激光雜志;2016年08期

2 陸建華;呂志才;;基于多時相點云數(shù)據(jù)的大型古建筑形變監(jiān)測研究——以蘇州虎丘塔為例[J];工程勘察;2016年07期

3 丁建勛;馬德富;高俊;;三維激光掃描數(shù)據(jù)精度影響因素分析及控制措施[J];城市勘測;2016年02期

4 張寧寧;楊英寶;于雙;;基于坡度和區(qū)域生長的城市LiDAR點云濾波方法[J];地理空間信息;2016年03期

5 徐進軍;廖驊;韓達(dá)光;邢誠;;大跨度橋梁橋面線形測量新方法[J];測繪通報;2016年01期

6 谷延超;范東明;余彪;張金花;;基于形態(tài)學(xué)與區(qū)域生長的機載LiDAR點云數(shù)據(jù)濾波[J];大地測量與地球動力學(xué);2015年05期

7 朱寧寧;;三維激光掃描在地鐵隧道形變監(jiān)測中的應(yīng)用[J];測繪工程;2015年05期

8 李濱;王國章;馮振;王文沛;;地下采空誘發(fā)陡傾層狀巖質(zhì)斜坡失穩(wěn)機制研究[J];巖石力學(xué)與工程學(xué)報;2015年06期

9 孫美玲;李永樹;陳強;蔡國林;;基于迭代多尺度形態(tài)學(xué)開重建的城區(qū)LiDAR濾波方法[J];紅外與激光工程;2015年01期

10 曹先革;張隨甲;司海燕;孟祥來;劉妍;;地面三維激光掃描點云數(shù)據(jù)精度影響因素及控制措施[J];測繪工程;2014年12期

相關(guān)會議論文 前1條

1 徐萬鑫;許宏麗;;改進的ICP算法在點云配準(zhǔn)中的應(yīng)用[A];Proceedings of 14th Chinese Conference on System Simulation Technology & Application(CCSSTA’2012)[C];2012年

相關(guān)碩士學(xué)位論文 前2條

1 陳衛(wèi);基于TLS技術(shù)的危巖體變形監(jiān)測方法研究[D];長安大學(xué);2014年

2 葛源坤;基于曲率特征信息的散亂點云數(shù)據(jù)預(yù)處理技術(shù)研究[D];西南交通大學(xué);2012年

，

本文編號：2022351