Point Cloud Filtering Research Articles

Vision‐Based Three‐Dimensional Reconstruction and Monitoring of Large‐Scale Steel Tubular Structures

A four‐ocular vision system is proposed for the three‐dimensional (3D) reconstruction of large‐scale concrete‐filled steel tube (CFST) under complex testing conditions. These measurements are vitally important for evaluating the seismic performance and 3D deformation of large‐scale specimens. A four‐ocular vision system is constructed to sample the large‐scale CFST; then point cloud acquisition, point cloud filtering, and point cloud stitching algorithms are applied to obtain a 3D point cloud of the specimen surface. A point cloud correction algorithm based on geometric features and a deep learning algorithm are utilized, respectively, to correct the coordinates of the stitched point cloud. This enhances the vision measurement accuracy in complex environments and therefore yields a higher‐accuracy 3D model for the purposes of real‐time complex surface monitoring. The performance indicators of the two algorithms are evaluated on actual tasks. The cross‐sectional diameters at specific heights in the reconstructed models are calculated and compared against laser rangefinder data to test the performance of the proposed algorithms. A visual tracking test on a CFST under cyclic loading shows that the reconstructed output well reflects the complex 3D surface after correction and meets the requirements for dynamic monitoring. The proposed methodology is applicable to complex environments featuring dynamic movement, mechanical vibration, and continuously changing features.

Improved Iterative Closest Point(ICP) 3D point cloud registration algorithm based on point cloud filtering and adaptive fireworks for coarse registration

ABSTRACTAiming at the problem of long computation time and poor registration accuracy in the current three-dimensional point cloud registration problem, this paper presents a k-dimensional Tree(KD-tree) improved ICP algorithm(KD-tree_ICP) that combines point cloud filtering and adaptive fireworks algorithms for coarse registration. On the basis of the typical KD-tree improved ICP algorithm, the point cloud filtering process and adaptive firework coarse registration process are added. Firstly, the point cloud data acquired by the 3D laser scanner is filtered. And then the adaptive fireworks algorithm is used to perform coarse registration on the filtered point cloud data. Next, the KD-tree_ICP algorithm is used to perform accurate registration on the basis of coarse registration, and the obtained translation and rotation relations are applied to the original point cloud data to obtain the result after registration. Finally, 3D point clouds of physical models of five statues are used for experimental verification, including error analysis, stability analysis and comparison with other algorithms. The experimental results show that the method proposed in this paper has greatly improved the calculation speed and accuracy, and the algorithm is stable and reliable, which can also be applied to the reconstruction of 3D building models, restoration of cultural relics, precision machining and other fields.

A compensation filter method for extracting surface characteristic from an optical point cloud

Optical devices cannot directly obtain the deformation for welding residual stress (WRS) measurement since it is difficult to get the surface characteristic with a high accuracy of measurement. This paper therefore presents a full-field fast method for WRS measurement based on heterodyne multiple frequency phase shift technology. Furthermore, in order to denoise the acquired point cloud, this paper proposes a novel compensation filter method (CFM) to extract accurate surface topography: specifically, to slice the original point cloud into several independent groups, then smooth each group of the point cloud to obtain the compensation value, and finally to fuse the filtered point cloud with the compensation value to extract surface characteristic. As a result, the simulation and the experiment results show that the CFM can reduce the noise by about 80%. It demonstrates that the novel CFM can maintain the features of the point cloud and remove noise efficiently, indicating the high accuracy of the CFM result and the potential to realize fast WRS measurement.

AUTOMATIC EXTRACTION OF POWER LINES FROM UAV LIDAR POINT CLOUDS USING A NOVEL SPATIAL FEATURE

Abstract. UAV LiDAR systems have unique advantage in acquiring 3D geo-information of the targets and the expenses are very reasonable; therefore, they are capable of security inspection of high-voltage power lines. There are already several methods for power line extraction from LiDAR point cloud data. However, the existing methods either introduce classification errors during point cloud filtering, or occasionally unable to detect multiple power lines in vertical arrangement. This paper proposes and implements an automatic power line extraction method based on 3D spatial features. Different from the existing power line extraction methods, the proposed method processes the LiDAR point cloud data vertically, therefore, the possible location of the power line in point cloud data can be predicted without filtering. Next, segmentation is conducted on candidates of power line using 3D region growing method. Then, linear point sets are extracted by linear discriminant method in this paper. Finally, power lines are extracted from the candidate linear point sets based on extension and direction features. The effectiveness and feasibility of the proposed method were verified by real data of UAV LiDAR point cloud data in Sichuan, China. The average correct extraction rate of power line points is 98.18%.

Synthesis of Leaf-on and Leaf-off Unmanned Aerial Vehicle (UAV) Stereo Imagery for the Inventory of Aboveground Biomass of Deciduous Forests

Applications of stereo imagery acquired by cameras onboard unmanned aerial vehicles (UAVs) as practical forest inventory tools are hindered by the unavailability of ground surface elevation. It is still a challenging issue to remove the elevation of ground surface in leaf-on stereo imagery to extract forest canopy height without the help of lidar data. This study proposed a method for the extraction of forest canopy height through the synthesis of UAV stereo imagery of leaf-on and leaf-off, and further demonstrated that the extracted forest canopy height could be used for the inventory of deciduous forest aboveground biomass (AGB). The points cloud of the leaf-on and leaf-off stereo imagery was firstly extracted by an algorithm of structure from motion (SFM) using the same ground control points (GCP). The digital surface model (DSM) was produced by rasterizing the point cloud of UAV leaf-on. The point cloud of UAV leaf-off was processed by iterative median filtering to remove vegetation points, and the digital terrain model (DTM) was generated by the rasterization of the filtered point cloud. The mean canopy height model (MCHM) was derived from the DSM subtracted by the DTM (i.e., DSM-DTM). Forest AGB maps were generated using models developed based on the MCHM and sampling plots of forest AGB and were evaluated by those of lidar. Results showed that forest AGB maps from UAV stereo imagery were highly correlated with those from lidar data with R2 higher than 0.94 and RMSE lower than 10.0 Mg/ha (i.e., relative RMSE 18.8%). These results demonstrated that UAV stereo imagery could be used as a practical inventory tool for deciduous forest AGB.

Photogrammetric surveying forests and woodlands with UAVs: techniques for automatic removal of vegetation and digital terrain model production for hydrological applications

The purpose of this study is the photogrammetric survey of a forested area using unmanned aerial vehicles (UAV), and the estimation of the digital terrain model (DTM) of the area, based on the photogrammetrically produced digital surface model (DSM). Furthermore, through the classification of the height difference between a DSM and a DTM, a vegetation height model is estimated, and a vegetation type map is produced. Finally, the generated DTM was used in a hydrological analysis study to determine its suitability compared to the usage of the DSM. The selected study area was the forest of Seih-Sou (Thessaloniki). The DTM extraction methodology applies classification and filtering of point clouds, and aims to produce a surface model including only terrain points (DTM). The method yielded a DTM that functioned satisfactorily as a basis for the hydrological analysis. Also, by classifying the DSM–DTM difference, a vegetation height model was generated. For the photogrammetric survey, 495 aerial images were used, taken by a UAV from a height of ∼200 m. A total of 44 ground control points were measured with an accuracy of 5 cm. The accuracy of the aerial triangulation was approximately 13 cm. The produced dense point cloud, counted 146 593 725 points.

3D point cloud filtering method for pose measurement application of space non-cooperative targets

3D point cloud filtering method for pose measurement application of space non-cooperative targets

DEM refinement by low vegetation removal based on the combination of full waveform data and progressive TIN densification

Filtering of low vegetation with height less than approximately 1.5 m is a challenging problem, especially in mountainous areas covered by heavy low foliage, bushes and sub-shrubberies, etc. The paper proposes an approach for obtaining a more accurate Digital Elevation Model (DEM) by removing low vegetation from point cloud. The approach combines point cloud with full waveform data, and begins by filtering point cloud by way of progressive TIN densification (PTD) method. Ground points are thus extracted, but mixed with false ground points, which are mainly from low vegetation and other manmade low objects. Gaussian decomposition by grouping Levenberg–Marquardt (LM) algorithm with F test is performed for the full waveforms corresponding to the extracted ground points. Echo widths and backscattering coefficients are calculated based on the parameters extracted from the decomposition, and used to discriminate points of low vegetation from points of other low objects, allowing the false ground points reflected from low vegetation to be labeled. New elevation values are calculated from the last echoes of the waveforms from low vegetation, and the DEM is updated by replacing the original elevations with the calculated ones. The resultants are assessed both quantitatively by check points and qualitatively by rendered DEM and contour lines generated from it. The accuracy of the refined DEM with low vegetation removal increases by 31% compared with the original DEM in the experiment, showing the effectiveness of the proposed approach.

Point cloud filtering on UAV based point cloud

Nowadays, Unmanned Aerial Vehicles (UAVs) have been attracted wide attentions such as a new measurement equipment and mapping, which are capable of the high-resolution point cloud data collection. In addition, a massive point cloud data has brought about the data filtering and irregular data organization for the generation of digital terrain models. Filtering of point clouds contains vegetations and artificial objects play a crucial role for bare earth terrain modelling. Topographical maps rely on the data structures which are built on bare ground terrain points. The bare earth surface extraction is not the only crucial to the topographical maps but also decision-making processes such as natural hazards management, deformation analysis and interpretation.In order to filter a UAV-based 3D raw point cloud data, in this paper, filtering performance of four different algorithms using open source and commercial software’s have been investigated, (1) curvature based (Multiscale Curvature Classification-MCC), (2) surface-based filtering (FUSION), (3) progressive TIN based (LasTool-LasGround module-commercial) and (4) physical simulation processing (Cloth Simulation Filtering-CSF). The applied filtering results were validated with the reference data set classified by operator. Although different filtering methodologies implemented on point clouds, these methods demonstrated similar results to extract ground on distinctive terrain feature such as dense vegetated, flat surface, rough and complex landscapes. The filtering algorithms’ results revealed that UAV-generated data suitable for extraction of bare earth surface feature on the different type of a terrain. Accuracy of the filtered point cloud reached the 93% true classification on flat surfaces from CSF filtering method.

Comparison of a commercial and home-assembled fixed-wing UAV for terrain mapping of a post-mining site under leaf-off conditions

ABSTRACTUnmanned aerial vehicle (UAV) platforms are rapidly becoming popular in many research and industry sectors. Due to their relatively low purchase price and the fact they can be used to monitor areas that are difficult or even unsafe to access, they have been increasingly used in land surveying and mapping of smaller areas. Numerous UAV platforms equipped with various cameras are increasingly available on the market, differing in their suitability for environmental mapping. Surveyors therefore face a question whether to buy or assemble their own UAV. The objective of this study is to assess the performance of two fixed-wing UAV systems for land survey and mapping applications. In particular, we: (1) compared a commercial eBee platform equipped with a Sony Cybershot DSC-WX220 camera with zoom lens and a home assembled EasyStar II equipped with Nikon Coolpix A with a lens of fixed focal length to find out if a home-assembled solution can compete with specialized commercial platform; (2) investigated the utilization of UAV images acquired under leaf-off conditions for digital terrain model (DTM) generation with respect to vegetation cover (steppes and forests); (3) assessed whether an increase in the image quantity can compensate for a lower quality of images; and (4) compared the DTM derived from UAV imagery with the official Czech Republic airborne laser scanning (ALS)-derived DTM. One flight with Easystar II and two perpendicular flights with eBee were performed. From these three flights, four point clouds were derived (one from each flight, and one resulting from a combination of two eBee flights), supplemented with four ground filtered point clouds. The accuracy of point clouds and DTM was assessed through a comparison with a conventional GNSS survey. We successfully identified the bare ground during the leaf-off period in the deciduous forest using images from both platforms. Point densities of point clouds acquired with Easystar II exceeded the densities of those acquired with eBee even after combining images from two eBee flights. Root mean square error of all derived point clouds ranged between 0.11 and 0.19 m, exceeding the accuracy of a nationwide ALS-derived DTM in both forest and open steppe areas. The most accurate point cloud was acquired using Easystar II. This is likely due to a combined effect of the quality of onboard cameras, camera settings and environmental conditions during the flight. For users who prefer to have greater control over their options rather than being dependent on the commercially available kit solution, home-assembled kits utilizing drones capable of carrying any camera available on the market may be an advantage.

Window Detection from UAS-Derived Photogrammetric Point Cloud Employing Density-Based Filtering and Perceptual Organization

Point clouds with ever-increasing volume are regular data in 3D city modelling, in which building reconstruction is a significant part. The photogrammetric point cloud, generated from UAS (Unmanned Aerial System) imagery, is a novel type of data in building reconstruction. Its positive characteristics, alongside its challenging qualities, provoke discussions on this theme of research. In this paper, patch-wise detection of the points of window frames on facades and roofs are undertaken using this kind of data. A density-based multi-scale filter is devised in the feature space of normal vectors to globally handle the matter of high volume of data and to detect edges. Color information is employed for the downsized data to remove the inner clutter of the building. Perceptual organization directs the approach via grouping and the Gestalt principles, to segment the filtered point cloud and to later detect window patches. The evaluation of the approach displays a completeness of 95% and 92%, respectively, as well as a correctness of 95% and 96%, respectively, for the detection of rectangular and partially curved window frames in two big heterogeneous cluttered datasets. Moreover, most intrusions and protrusions cannot mislead the window detection approach. Several doors with glass parts and a number of parallel parts of the scaffolding are mistaken as windows when using the large-scale object detection approach due to their similar patterns with window frames. Sensitivity analysis of the input parameters demonstrates that the filter functionality depends on the radius of density calculation in the feature space. Furthermore, successfully employing the Gestalt principles in the detection of window frames is influenced by the width determination of window partitioning.

A Gaussian filtering method to reduce directionality on high-density point clouds digitized by a conoscopic holography sensor

This work analyses the directional effect shown by the high-density point clouds digitized with a conoscopic holography (CH) sensor. The asymmetric shape of the laser spot for this sensor yields directionality to appear along the largest spot dimension and to occur repeatedly under different working conditions. To study this effect, several digitizing tests were performed under different conditions on a surface machined by EDM with a uniform and isotropic finish, so that the directional effect should not appear. Nevertheless, the use of the 2D Fourier transform (2DFT) confirmed the existence of directionality in the point clouds along the largest spot direction and that it appeared repetitively under different working conditions. Thus, this effect could be considered as a systematic error associated to the CH sensor and then, feasible to be reduced. The use of an anisotropic 2D Gaussian filter is suggested for this purpose. The results found before and after applying the filter were compared to those obtained by a confocal microscope, which was used as reference due to the absence of directionality in the captured images. Results show that the filtered point clouds suitably fit the actual surface topography.

Multi-Normal Estimation via Pair Consistency Voting.

The normals of feature points, i.e., the intersection points of multiple smooth surfaces, are ambiguous and undefined. This paper presents a unified definition for point cloud normals of feature and non-feature points, which allows feature points to possess multiple normals. This definition facilitates several succeeding operations, such as feature points extraction and point cloud filtering. We also develop a feature preserving normal estimation method which outputs multiple normals per feature point. The core of the method is a pair consistency voting scheme. All neighbor point pairs vote for the local tangent plane. Each vote takes the fitting residuals of the pair of points and their preliminary normal consistency into consideration. Thus the pairs from the same subspace and relatively far off features dominate the voting. An adaptive strategy is designed to overcome sampling anisotropy. In addition, we introduce an error measure compatible with traditional normal estimators, and present the first benchmark for normal estimation, composed of 152 synthesized data with various features and sampling densities, and 288 real scans with different noise levels. Comprehensive and quantitative experiments show that our method generates faithful feature preserving normals and outperforms previous cutting edge normal estimation methods, including the latest deep learning based method.

An automated algorithm for mapping building impervious areas from airborne LiDAR point-cloud data for flood hydrology

Buildings, as impervious surfaces, are an important component of total impervious surface areas that drive urban stormwater response to intense rainfall events. Most stormwater models that use percent impervious area (PIA) are spatially lumped models and do not require precise locations of building roofs, as in other applications of building maps, but do require accurate estimates of total impervious areas within the geographic units of observation (e.g. city blocks or sub-watershed units). Two-dimensional mapping of buildings from aerial imagery requires laborious efforts from image analysts or elaborate image analysis techniques using high spatial resolution imagery. Moreover, large uncertainties exist where tall, dense vegetation obscures the structures. Analyzing LiDAR point-cloud data, however, can distinguish buildings from vegetation canopy and facilitate the mapping of buildings. This paper presents a new building extraction approach that is based on and optimized for estimating building impervious areas (BIA) for hydrologic purposes and can be used with standard GIS software to identify building roofs under tall, thick canopy. Accuracy assessment methods are presented that can optimize model performance for modeling BIA within the geographic units of observation for hydrologic applications. The Building Extraction from LiDAR Last Returns (BELLR) model, a 2.5D rule-based GIS model, uses a non-spatial, local vertical difference filter (VDF) on LiDAR point-cloud data to automatically identify and map building footprints. The model includes an absolute difference in elevation (AdE) parameter in the VDF that compares the difference between mean and modal elevations of last-returns in each cell.The BELLR model is calibrated for an extensive inner-city, highly urbanized small watershed in Columbia, South Carolina, USA that is covered by tall, thick vegetation canopy that obscures many buildings. The calibration of BELLR used a set of building locations compiled by photo-analysts, and validation used independent building reference data. The model is applied to two residential neighborhoods, one of which is a residential area within the primary watershed and the other is a younger suburban neighborhood with a less-well developed tree canopy used as a validation site. Performance results indicate that the BELLR model is highly sensitive to concavity in the lasboundary tool of LAStools® and those settings are highly site specific. The model is also sensitive to cell size and the AdE threshold values. However, properly calibrated the BIA for the two residential sites could be estimated within 1% error for optimized experiments.To examine results in a hydrologic application, the BELLR estimated BIAs were tested using two different types of hydrologic models to compare BELLR results with results using the National Land Cover Database (NLCD) 2011 Percent Developed Imperviousness data. The BELLR BIA values provide more accurate results than the use of the 2011 NLCD PIA data in both models. The VDF developed in this study to map buildings could be applied to LiDAR point-cloud filtering algorithms for feature extraction in machine learning or mapping other planar surfaces in more broad-based land-cover classifications.

Field Navigation Using Fuzzy Elevation Maps Built with Local 3D Laser Scans

The paper describes the case study of the mobile robot Andabata navigating on natural terrain at low speeds with fuzzy elevation maps (FEMs). To this end, leveled three-dimensional (3D) point clouds of the surroundings are obtained by synchronizing ranges obtained from a 360 ∘ field of view 3D laser scanner with odometric and inertial measurements of the vehicle. Then, filtered point clouds are employed to produce FEMs and their corresponding fuzzy reliability masks (FRMs). Finally, each local FEM and its FRM are processed to choose the best motion direction to reach distant goal points through traversable areas. All these tasks have been implemented using ROS (Robot Operating System) nodes distributed among the cores of the onboard processor. Experimental results of Andabata during non-stop navigation on an urban park are presented.

Study of rapid face modeling technology based on Kinect

This paper improves the algorithm of point cloud filtering and registration in 3D modeling, aiming for smaller sampling error and shorter processing time of point cloud data. Based on collaborative sampling among several Kinect devices, we analyze the deficiency of current filtering algorithm, and use a novel method of point cloud filtering. Meanwhile, we use Fast Point Feature Histogram (FPFH) algorithm for feature extraction and point cloud registration. Compared with the aligning process using Point Feature Histograms (PFH), it only takes 9[Formula: see text]min when the number of points is about 500,000, shortening the aligning time by 47.1%. To measure the accuracy of the registration, we propose an algorithm which calculates the average distance of the corresponding coincident parts of two point clouds, and we improve the accuracy to an average distance of 0.7[Formula: see text]mm. In the surface reconstruction section, we adopt Ball Pivoting algorithm for surface reconstruction, obtaining image with higher accuracy in a shorter time.

机载LiDAR点云滤波综述

机载LiDAR点云滤波综述

Guided 3D point cloud filtering

3D point cloud has gained significant attention in recent years. However, raw point clouds captured by 3D sensors are unavoidably contaminated with noise resulting in detrimental efforts on the practical applications. Although many widely used point cloud filters such as normal-based bilateral filter, can produce results as expected, they require a higher running time. Therefore, inspired by guided image filter, this paper takes the position information of the point into account to derive the linear model with respect to guidance point cloud and filtered point cloud. Experimental results show that the proposed algorithm, which can successfully remove the undesirable noise while offering better performance in feature-preserving, is significantly superior to several state-of-the-art methods, particularly in terms of efficiency.

A Multi-Constraint Combined Method for Ground Surface Point Filtering from Mobile LiDAR Point Clouds

Point cloud filtering is an essential preprocessing step in 3D (three-dimensional) LiDAR (light detection and ranging) point cloud processing. The filtering of mobile LiDAR scanning point clouds is much more challenging due to their non-uniform distribution, the large-scale of missing data areas and the existence of both large size objects and small land features. This paper proposes a new filtering method that combines range constraint, slope constraint and angular position constraint to filter ground surface points from mobile LiDAR point clouds. Firstly, a cylindrical coordinate system (CCS) is established for each block of mobile LiDAR point clouds and three attributes of mobile LiDAR points, i.e., the angular position attribute (AA), longitudinal distance attribute (LA) and range attribute (RA), are computed. Then, the mobile LiDAR point clouds are structured into a grid according to the AA and LA. Finally, the point clouds are filtered by a single cross-section filter (SCSF) using range constraint and slope constraint, followed by a multiple cross-section filter (MCSF) using range constraint and angular position constraint. Five datasets are used to validate the proposed method. The experimental results show that the proposed new filtering method achieves an average type I error, type II error, and total error of 1.426%, 1.885%, and 1.622%, respectively.

Creating high-resolution bare-earth digital elevation models (DEMs) from stereo imagery in an area of densely vegetated deciduous forest using combinations of procedures designed for lidar point cloud filtering

For areas of the world that do not have access to lidar, fine-scale digital elevation models (DEMs) can be photogrammetrically created using globally available high-spatial resolution stereo satellite imagery. The resultant DEM is best termed a digital surface model (DSM) because it includes heights of surface features. In densely vegetated conditions, this inclusion can limit its usefulness in applications requiring a bare-earth DEM. This study explores the use of techniques designed for filtering lidar point clouds to mitigate the elevation artifacts caused by above ground features, within the context of a case study of Prince William Forest Park, Virginia, USA. The influences of land cover and leaf-on vs. leaf-off conditions are investigated, and the accuracy of the raw photogrammetric DSM extracted from leaf-on imagery was between that of a lidar bare-earth DEM and the Shuttle Radar Topography Mission DEM. Although the filtered leaf-on photogrammetric DEM retains some artifacts of the vegetation canopy and may not be useful for some applications, filtering procedures significantly improved the accuracy of the modeled terrain. The accuracy of the DSM extracted in leaf-off conditions was comparable in most areas to the lidar bare-earth DEM and filtering procedures resulted in accuracy comparable of that to the lidar DEM.