基于距离限制的机载激光数据滤波方法

提出了基于距离限制的滤波算法，利用地物在三维空间的距离特性提取地面信息，并对可能影响算法稳健性的侧面信息提出角度判决的概念。试验表明，本文方法在处理复杂城市地形时，能很好地保留地面的细节信息，剔除低矮植被等非地面信息。     

LiDAR测量点云融合影像的分块滤波方法

针对现有LiDAR地面点滤波算法对复杂地形地物适应性不强的问题,本文提出了一种融合点云与地面影像分块滤波的方法。首先,将地面影像与点云匹配,使点云从影像中获取更多的光谱纹理信息。然后,分析地物光谱、林地相对密度、点云高程特征、地面DSM模型及其坡度,并基于决策级融合将原始点云切割成若干独立的区块。最后,根据每块区域不同的多元细节特征,对IPTD滤波算法进行改进并利用搜索法优化参数,得到最优且稳健的结果。利用滤波后的总地面点通过插值算法得到的DEM模型和相关试验验证了本文算法的优越性。     

基于车载激光点云的高精地图矢量化成图

针对车载激光点云中对各特征物提取结果后矢量化成图时的自动化问题,本文基于双方向积分法实现了边缘检测及矢量化成图,旨在保证特征物基本特征的同时,也保证点云的绝对精度。将输入的特征点云进行离群点过滤,以保证外包框算法特征点云的准确度;将三维点云按照外包框算法投影至最优平面,为后续沿各方向积分提供输入;利用八邻域KD-tree算法求出样本特征点云的均值邻域半径,依据邻域半径对各方向积分提供积分域中的微分元;根据提供的微分元沿各方向进行积分,在该积分元内找到距平面最值的最优解;按照积分结果构建点云索引,并根据点云特征构建模型,最终得到高精地图的矢量化点云。试验证明了该方法在处理实际问题时的可行性。     

一种改进的超体素与区域生长点云分割方法

点云分割作为识别地理场景的空间特征、探索和记录空间信息的关键处理步骤,其分割精度直接影响后续三维场景重建、地物特征提取等应用的效果。针对传统区域生长点云分割算法的不稳定等问题,本文结合超体素和区域生长算法对点云数据进行分割,并利用点云自身的色彩信息进一步改进分割结果。试验结果表明,相较于传统的区域生长和已有的结合超体素与区域生长的分割算法,本文算法对点云数据分割的效果更好,且其精确率与召回率均有提高。     

Pattern detection in airborne LiDAR data using Laplacian of Gaussian filter

Methods for feature detection in laser scanning data have been studied for decades ever since the emergence of the technology.However,it is still one of the unsolved problems in LiDAR data processing due to difficulty of texture and structure information extraction in unevenly sampled points.The paper analyzes the characteristics of Laplacian of Gaussian(LoG) Filter and its potential use for structure detection in LiDAR data.A feature detection method based on LoG filtering is presented and ex-perimented on the unstructured points.The method filters the elevation value(namely,z coordinate value) of each point by convo-lution using LoG kernel within its local area and derives patterns suggesting the existence of certain types of ground ob-jects/features.The experiments are carried on a point cloud dataset acquired from a neighborhood area.The results demonstrate patterns detected at different scales and the relationship between standard deviation that defines LoG kernel and neighborhood size,which specifies the local area that is analyzed.     

Distance-limited filter for extracting ground points

The extraction of points on the bare earth from point clouds acquired by airborne laser scanning is one of the most important steps for the generation of digital terrain models (DTM). This process is called “filtering”. However, most of the current filters erode the bare earth in steep sloped landscapes and at discontinuities, and they retain low vegetation. Therefore, a new filtering method for extracting ground points based on a distance limit is proposed in this paper. The angle criterion is used to assure the robustness of the algorithm. The experimental results show that the proposed filtering method can effectively derive the ground points from point clouds in complex urban areas.     

基于散乱点云特征提取算法研究

三维激光扫描测量技术是当前测绘领域研究的热点。点云各个要素特征提取在三维建模中是非常重要的一个环节,也逐渐成为三维点云数据处理中的一个研究重点。目前对点云数据特征提取仍然存在一些问题,大多数算法的研究是针对栅格点云数据进行的,存在容易受到噪声数据的影响以及适应性不强等问题。本文在研究了现有的点云数据特征提取的方法的基础上,将法线差分算法应用到场景内地物提取中,从而实现场景中地物的特征提取。     

一种基于分割的机载LiDAR点云数据滤波

针对当前滤波算法在处理地形不连续区域或存在复杂建筑物区域时容易过分＂腐蚀＂地形并难以去除一些低矮植被的不足,提出了一种基于分割的机载LiDAR点云滤波算法。首先,对原始点云基于地表连续性进行分割;然后,在移除点数目较小的粗差点集之后采用对分割点集建立缓冲区的方法,区分地面和非地面点集;在较大地物经过迭代分割基本移除之后,使用约束平面的方法移除高度较小的地表附着物以实现滤波。实验结果表明,与经典滤波算法相比,该算法提高了地面点的分类精度,在滤除地物信息的同时能有效地保留地形特征。     

Pattern detection in airborne LiDAR data using Laplacian of Gaussian filter

Methods for feature detection in laser scanning data have been studied for decades ever since the emergence of the technology. However, it is still one of the unsolved problems in LiDAR data processing due to difficulty of texture and structure information extraction in unevenly sampled points. The paper analyzes the characteristics of Laplacian of Gaussian (LoG) Filter and its potential use for structure detection in LiDAR data. A feature detection method based on LoG filtering is presented and experimented on the unstructured points. The method filters the elevation value (namely, z coordinate value) of each point by convolution using LoG kernel within its local area and derives patterns suggesting the existence of certain types of ground objects/features. The experiments are carried on a point cloud dataset acquired from a neighborhood area. The results demonstrate patterns detected at different scales and the relationship between standard deviation that defines LoG kernel and neighborhood size, which specifies the local area that is analyzed.     

基于深度残差网络的机载LiDAR点云分类

机载LiDAR点云的分类是利用其进行城市场景三维重建的关键步骤之一。为充分利用现有的图像领域性能较好的深度学习网络模型,提高点云分类精度,并降低训练时间和对训练样本数量的要求,本文提出一种基于深度残差网络的机载LiDAR点云分类方法。首先提取归一化高程、表面变化率、强度和归一化植被指数4种具有较高区分度的点云低层次特征;然后通过设置不同的邻域大小和视角,利用所提出的点云特征图生成策略,得到多尺度和多视角点云特征图;再将点云特征图输入到预训练的深度残差网络,提取多尺度和多视角深层次特征;最后构建并训练神经网络分类器,利用训练的模型对待分类点云进行预测,经后处理得到分类结果。利用ISPRS三维语义标记竞赛的公开标准数据集进行试验,结果表明,本文方法可有效区分建筑物、地面、车辆等8类地物,分类结果的总体精度为87.1%,可为城市场景三维重建提供可靠的信息。     

面向水流三维登记探索的多波束点云去噪及应用

针对多波束点云数据去噪难以保留精细特征,无法精确“锁定”河床形态问题,本文提出了一种面向自然资源确权水流三维登记探索的多波束点云去噪算法。以KD树搜索为基础,引入统计滤波理论进行多尺度噪声分类,并剔除大尺度噪声;针对小尺度噪声,在信息熵理论基础上,以主成分分析算法为基础,以信息熵最小原则确定最优邻域,并据此构建曲率信息熵对双边滤波因子进行优化改进,以实现水下地形点云去噪与精细特征保留的目的。试验结果表明,本文算法具有可行性,能够有效保证水下地形的精细特征,并能够应用于自然资源水流三维登记。     

基于距离限制的地面点提取滤波器(英文)

The extraction of points on the bare earth from point clouds acquired by airborne laser scanning is one of the most important steps for the generation of digital terrain models (DTM). This process is called “filtering”. However, most of the current filters erode the bare earth in steep sloped landscapes and at discontinuities, and they retain low vegetation. Therefore, a new filtering method for extracting ground points based on a distance limit is proposed in this paper. The angle criterion is used to assure the robustness of the algorithm. The experimental results show that the proposed filtering method can effectively derive the ground points from point clouds in complex urban areas. Supported by the Program for Changjiang Scholars and Innovative Research Team in University (No.0438), the National 863 Program of China (No. 2006AA12Z151).     

建筑物立面分类与DIM点云滤波

针对目前基于近景摄影测量方法构建建筑物立面模型过程中因密集影像匹配（DIM）点云噪声所引起的建筑物立面TIN网格模型畸变问题,本文借鉴机器学习中样本学习的思想,对建筑物立面进行了分类并对DIM点云提出了相应的滤波方法,以达到去除DIM点云噪声和改善其TIN网格模型畸变的目的。其中,针对平面结构立面,采取先对点云样本进行学习计算构建数学立面模型所需参数,再对该立面模型设定阈值并对其点云进行滤波处理的方法;针对曲面结构立面,则结合DIM点云特性先将点云样本分类标记归为立面点与非立面点,再进行样本特征值学习,使用Logistic回归算法迭代计算求解最佳回归系数,从而构建滤波分类器的方法对立面点云进行滤波处理。试验结果表明,本文滤波处理方法能将立面DIM点云噪声有效识别并去除,而且使用该方法处理后所得点云构建的建筑物立面TIN网格模型精细化程度得到有效提高,模型质量得到明显改善。     

TERRA: Terrain Extraction from elevation Rasters through Repetitive Anisotropic filtering

Over the past decades, several filters have been developed to derive a Digital Terrain Model (DTM) from a Digital Surface Model (DSM), by means of filtering out aboveground objects such as vegetation. In this filtering process, however, one of the major challenges remains to precisely distinguish sharp terrain features, e.g. ridges, agricultural terraces or other anthropogenic geomorphology such as open-pit mines, riverbanks or road ramps. Hence, loss of elevation data around terrain edges (and consequent smoothing) is very common with existing algorithms. In terraced landscapes, the preservation of precise geomorphology is of key importance in digital terrain analyses, such as hydrologic and erosion modelling, or automatic feature recognition and inventorying. In this work, we propose a new filtering method called TERRA (Terrain Extraction from elevation Rasters through Repetitive Anisotropic filtering). The novelty of the algorithm lies within its usage of terrain aspect to guide the anisotropic filtering direction, therefore maximising the preservation of terrain edges. We derived six DTMs from DSMs using UAV Structure from Motion (SfM) photogrammetry, laser altimetry and satellite sources (grid resolutions ranging from 0.1–1.0 m). The results indicated a close agreement of DTMs filtered using the TERRA algorithm and reference DTMs, while terrace risers were well preserved even under thick canopies of vines and trees. Compared to existing filtering approaches, TERRA performed well in minimising Type I errors (false ground removal), while Type II errors occurred locally where vegetation was covering the terrace edges. Given the promising filtering performance, and supported by the minimal requirements of parameterisation and computation, the TERRA algorithm could be a useful tool in DTM preparation for digital terrain analysis of agricultural terraces and similar hillslopes characterised by a complex mosaic of sharp terrain and non-terrain features.     

深度图像分割的城市区域倾斜影像密集匹配点云滤波算法

倾斜摄影测量作为一个新兴领域发展势头迅猛,在众多领域得到了广泛应用。但倾斜影像密集匹配点云处理技术研究却相对较少。倾斜影像密集匹配点云分布不均匀、表面粗糙,因而传统的激光扫描点云处理算法在用于倾斜影像密集匹配点云处理时的适用性较低。本文从倾斜影像密集匹配点云特点出发,提出了一种利用点云高程信息生成深度图像提取建筑物非连通区域,在全局范围选取种子点实现多种子点区域生长的点云快速滤波算法。实验结果表明,该算法滤波效果好、速度快,可以改善密集匹配点云部分地物底部边缘不清晰引起错分和区域生长无法分割建筑物非连通区域的问题。     

机载激光雷达点云滤波算法分析与比较

机载点云数据在城市三维建模、DEM提取中应用广泛,而机载点云滤波是这些应用的基础。因此,这里对机载点云滤波算法设计所依据的地面特征和滤波结果的精度评定方法作了总结,并对现有滤波算法进行了分类描述。最后着重对滤波算法做了直观的对比分析,为后续点云数据滤波处理研究提供参考。     

一种多尺度自适应点云坡度滤波算法

点云坡度滤波算法原理简单、易于实现,为进一步提升坡度滤波算法的自适应性,提出了一种多尺度自适应点云坡度滤波算法.首先,在数据预处理的基础上引入虚拟网格对点云数据进行分割;然后,利用距离加权的方式逐次计算网格点的坡度角,结合k均值聚类和正态分布自适应确定滤波阈值;最后,使用多尺度策略逐级缩小网格尺寸实现点云数据的精细滤波...     

Image classification-based ground filtering of point clouds extracted from UAV-based aerial photos

With the advent of unmanned aerial vehicles (UAVs) for mapping applications, it is possible to generate 3D dense point clouds using stereo images. This technology, however, has some disadvantages when compared to Light Detection and Ranging (LiDAR) system. Unlike LiDAR, digital cameras mounted on UAVs are incapable of viewing beneath the canopy, which leads to sparse points on the bare earth surface. In such cases, it is more challenging to remove points belonging to above-ground objects using ground filtering algorithms generated especially for LiDAR data. To tackle this problem, a methodology employing supervised image classification for filtering 3D point clouds is proposed in this study. A classified image is overlapped with the point cloud to determine the ground points to be used for digital elevation model (DEM) generation. Quantitative evaluation results showed that filtering the point cloud with this methodology has a good potential for high-resolution DEM generation.     

利用双向布料模拟的机载建筑物点云提取

针对现有机载激光扫描数据的建筑物提取方法过程复杂且易受植被干扰的问题,本文提出了一种利用双向布料模拟策略的建筑物提取方法。首先在正向布料模拟滤波的基础上,构建归一化数字表面模型提取过高建筑物,并采用反向布料模拟,从其余地物点中粗提取建筑物顶面点云;然后进行穿透性分析,并结合形态学操作进一步剔除错提的植被点;最后,以包含顶面点云的三维格网为种子格网,根据格网之间的邻接关系和内部点云几何特征进行约束生长,获取完整建筑物点云。试验结果表明,在复杂场景中,该方法能够有效避免植被的干扰,快速提取建筑物点云,具有提取精度高、计算时间少的优点。     

基于移动网格点云精简算法的研究

针对三维激光点云数据精简问题,提出一种基于移动网格划分的精简算法。基本思想是首先将点云模型空间网格化,再依据距离阈值对网格进行二次网格划分,利用权重值大小筛选出合适点,最后移动点云模型包围盒最小点位置,重新进行二次网格划分和点云筛选,最终完成点云的精简。实验得出,可移动网格划分的点云精简算法在精简效果与时间效率上均具有显著的优势。