Non-planar Points Research Articles

Research on Point Cloud Structure Detection of Manhole Cover Based on Structured Light Camera

This study introduced an innovative approach for detecting structural anomalies in road manhole covers using structured light cameras. Efforts have been dedicated to enhancing data quality by commencing with the acquisition and preprocessing of point cloud data from real-world manhole cover scenes. The RANSAC algorithm is subsequently employed to extract the road plane and determine the height of the point cloud structure. In the presence of non-planar point cloud exhibiting abnormal heights, the DBSCAN algorithm is harnessed for cluster segmentation, aiding in the identification of individual objects. The method culminates with the introduction of a sector fitting detection model, adept at effectively discerning manhole cover features within the point cloud and delivering comprehensive height and structural information. Experimental findings underscore the method’s efficacy in accurately gauging the degree of subsidence in manhole cover structures, with data errors consistently maintained within an acceptable range of 8 percent. Notably, the measurement speed surpasses that of traditional methods, presenting a notably efficient and dependable technical solution for road maintenance.

An adaptive down-sampling method of laser scan data for scan-to-BIM

Laser scan data are popularly adopted to capture the conditions of existing buildings for as-is BIM reconstruction, also known as scan-to-BIM. Down-sampling of massive laser scan data is a critical pre-processing step for efficient scan-to-BIM. It is also crucial to accurately retain the geometric and semantic information of existing buildings during the down-sampling process. However, existing down-sampling approaches mainly focus on the preservation of geometric features only, without considering semantic features. This study developed an adaptive down-sampling method, which is able to maintain scan points containing critical geometric or semantic information and only down-sample scan points containing no critical information. The proposed method first conducts a geometry-based segmentation to identify edge points and non-planar points, which contain critical geometric information. Then, a semantic-based segmentation is performed to identify points with critical semantic information (e.g. labels and information boards). All the points containing either geometric or semantic information are retained, while points containing no critical information are down-sampled to reduce data redundancy. Experiments were conducted on four scenes, which showed that the proposed method was suitable for point cloud data with different accuracies and could achieve a better performance in preserving both geometric and semantic information than the traditional voxel-based method and the curvature-based method. It is proved that down-sampled data with the proposed method could generate as-is BIMs with more accurate geometric and semantic information.

A GPU-Based Radio Wave Propagation Prediction With Progressive Processing on Point Cloud

Point cloud data (PCD) can record a high-resolution model of the environment much more efficiently than the conventional geometrical mesh. This feature makes PCD suitable for the radio channel prediction of a new environment, especially for the millimeter-wave (mmWave). Since the multiple propagations and visibility computation using PCD are time consuming, progressive processing on PCD is introduced to make the prediction compatible with GPU-based frameworks such as OptiX and CUDA. Compared with the surface reconstruction from PCD, progressive data from the processing reserves more domain features such as diffraction wedge and the label of planar or nonplanar points. The numerical result of an outdoor environment shows that the proposed method has a close agreement with measurement, as well as two impressive speedups, 49.8 for the ray tracing and field calculation and 18.8 for the total prediction, compared with the original method using PCD.

Clustering-Based Plane Refitting of Non-planar Patches for Voxel-Based 3D Point Cloud Segmentation Using K-Means Clustering

Point cloud processing is a struggled field because the points in the clouds are three-dimensional and irregular distributed signals. For this reason, the points in the point clouds are mostly sampled into regularly distributed voxels in the literature. Voxelization as a pretreatment significantly accelerates the process of segmenting surfaces. The geometric cues such as plane directions (normals) in the voxels are mostly used to segment the local surfaces. However, the sampling process may include a non-planar point group (patch), which is mostly on the edges and corners, in a voxel. These voxels can cause misleading the segmentation process. In this paper, we separate the non-planar patches into planar sub-patches using k-means clustering. The largest one among the planar sub-patches replaces the normal and barycenter properties of the voxel with those of itself. We have tested this process in a successful point cloud segmentation method and measure the effects of the proposed method on two point cloud segmentation datasets (Mosque and Train Station). The method increases the accuracy success of the Mosque dataset from 83.84% to 87.86% and that of the Train Station dataset from 85.36% to 87.07%.

Roof Plane Segmentation from Airborne LiDAR Data Using Hierarchical Clustering and Boundary Relabeling

The roof plane segmentation is one of the key issues for constructing accurate three-dimensional building models from airborne light detection and ranging (LiDAR) data. Region growing is one of the most widely used methods to detect roof planes. It first selects one point or region as a seed, and then iteratively expands to neighboring points. However, region growing has two problems. The first problem is that it is hard to select the robust seed points. The other problem is that it is difficult to detect the accurate boundaries between two roof planes. In this paper, to solve these two problems, we propose a novel approach to segment the roof planes from airborne LiDAR point clouds using hierarchical clustering and boundary relabeling. For the first problem, we first extract the initial set of robust planar patches via an octree-based method, and then apply the hierarchical clustering method to iteratively merge the adjacent planar patches belonging to the same plane until the merging cost exceeds a predefined threshold. These merged planar patches are regarded as the robust seed patches for the next region growing. The coarse roof planes are generated by adding the non-planar points into the seed patches in sequence using region growing. However, the boundaries of coarse roof planes may be inaccurate. To solve this problem, namely, the second problem, we refine the boundaries between adjacent coarse planes by relabeling the boundary points. At last, we can effectively extract high-quality roof planes with smooth and accurate boundaries from airborne LiDAR data. We conducted our experiments on two datasets captured from Vaihingen and Wuhan using Leica ALS50 and Trimble Harrier 68i, respectively. The experimental results show that our proposed approach outperforms several representative approaches in both visual quality and quantitative metrics.

Registration of partially overlapping laser-radar range images

To register partially overlapping three-dimensional point sets from different viewpoints, it is necessary to remove spurious corresponding point pairs that are not located in overlapping regions. Most variants of the iterative closest point (ICP) algorithm require users to manually select the rejection parameters for discarding spurious point pairs between the registering views. This requirement often results in unreliable and inaccurate registration. To overcome this problem, we present an improved ICP algorithm that can automatically determine the rejection percentage to reliably and accurately align partially overlapping laser-radar (ladar) range images. The similarity of k neighboring features of each nonplanar point is employed to determine reasonable point pairs in nonplanar regions, and the distance measurement method is used to find reasonable point pairs in planar regions. The rejection percentage can be obtained from these two sets of reasonable pairs. The performance of our algorithm is compared with that of five other algorithms using various models with low and high curvatures. The experimental results show that our algorithm is more accurate and robust than the other algorithms. © 2015

Classification-based scene modeling for urban point clouds

The three-dimensional modeling of urban scenes is an important topic that can be used for various applications. We present a comprehensive strategy to reconstruct a scene from urban point clouds. First, the urban point clouds are classified into the ground points, planar points on the ground, and nonplanar points on the ground by using the support vector machine algorithm which takes several differential geometry properties as features. Second, the planar points and nonplanar points on the ground are segmented into patches by using different segmentation methods. A collection of characteristics of point cloud segments like height, size, topological relationship, and ratio between the width and length are applied to extract different objects after removing the unwanted segments. Finally, the buildings, ground, and trees in the scene are reconstructed, resulting in a hybrid model representing the urban scene. Experimental results demonstrate that the proposed method can be used as a robust way to reconstruct the scene from the massive point clouds.

Seams of Conical Intersections Relevant to the Quenching of OH(A2Σ+) by Collisions with H2

Using multireference configuration interaction wave functions composed of 17–52 million configuration state functions, 18 points on the 1(2)A–2(2)A seam and 162 points on the 2(2)A–3(2)A seam of conical intersections relevant to the collisional quenching of OH(A(2)Σ(+)) by H(2) are determined and analyzed. In the vicinity of planar nuclear configurations, the former seam corresponds to a 1(2)A′–1(2)A″ seam of intersection and the latter corresponds to a 1(2)A′–2(2)A′ seam. For the previously studied 2(2)A–3(2)A seam, two regions not previously examined are reported: (i) an out-of-plane region that connects smoothly to the 1(2)A′–2(2)A′ seam for planar structures and (ii) a Rydberg region that includes D(3h)/C(3v) structures where the 2(2)A–3(2)A seam is a 2(2)A′–3(2)A′ seam for D(3h) structures. Some of the nonplanar points on the 2(2)A–3(2)A seam of conical intersection are found to have OH and H(2) distances comparable to those of the reactant molecules and energies below that of the reactant asymptote. These nonplanar entrance channel conical intersections suggest new mechanisms for the quenching reaction. The Rydberg region introduces new connectivity and symmetry issues. For the 1(2)A–2(2)A [1(2)A′–1(2)A″] seam, which unlike the 2(2)A–3(2)A [1(2)A′–2(2)A′] seam cannot continuously deform from planar to nonplanar structures except through confluences, no evidence of nonplanar points on the conical intersection seam was found. The continuous conical parameters, g(I,J), h(I,J), s(x)(I,J), and s(y)(I,J) and the associated vectors g(I,J), h(I,J), and s(I,J), are determined and discussed. The conical parameters are made continuous by a prescribed rotation of the degenerate wave functions. The continuity of these conical parameters makes it possible to construct a quasi-diabatic representation of the coupled adiabatic potential energy surfaces.

A novel plane extraction approach using supervised learning

This paper presents a novel approach for the classification of planar surfaces in an unorganized point clouds. A feature-based planner surface detection method is proposed which classifies a point cloud data into planar and non-planar points by learning a classification model from an example set of planes. The algorithm performs segmentation of the scene by applying a graph partitioning approach with improved representation of association among graph nodes. The planarity estimation of the points in a scene segment is then achieved by classifying input points as planar points which satisfy planarity constraint imposed by the learned model. The resultant planes have potential application in solving simultaneous localization and mapping problem for navigation of an unmanned-air vehicle. The proposed method is validated on real and synthetic scenes. The real data consist of five datasets recorded by capturing three-dimensional(3D) point clouds when a RGBD camera is moved in five different indoor scenes. A set of synthetic 3D scenes are constructed containing planar and non-planar structures. The synthetic data are contaminated with Gaussian and random structure noise. The results of the empirical evaluation on both the real and the simulated data suggest that the method provides a generalized solution for plane detection even in the presence of the noise and non-planar objects in the scene. Furthermore, a comparative study has been performed between multiple plane extraction methods.

UAV ATTITUDE ESTIMATION BY COMBINING HORIZON-BASED AND HOMOGRAPHY-BASED APPROACHES FOR CATADIOPTRIC IMAGES

Developing Unmanned Aerial Vehicles (UAV) with self-stabilization capabilities represents an intensive research field nowadays. This paper aims to show that combining our previous horizon extraction algorithm with homography methods permits to estimate the homography more robustly and thus the UAV attitude. We show that imposing horizon constraint permits to remove evident inconvenient non-planar points. Moreover, we explain that computing the horizon in the sphere permits to compute the normal vector of the ground plane and thus retrieve the right motion among the four possible solutions obtained by planar homography, which is not a trivial problem usually.

3-D object recognition using a new invariant relationship by single-view

We propose a new method for recognizing three-dimensional objects using a three-dimensional invariant relationship and geometric hashing by single-view. We develop a special structure consisting of four co-planar points and any two non-planar points with respect to the plane. We derive an invariant relationship for the structure, which is represented by a plane equation. For the recognition of three-dimensional objects using the geometric hashing, a set of points on the plane, thereby satisfying the invariant relationship, are mapped into a set of points intersecting the plane and the unit sphere. Since the structure is much more general than the previous structures proposed by Rothwell et al. (Oxford University TR-OUEL 1927/92, 1992) and Zhu et al. (Proceedings of the 12th International Conference on Robotics and Automation, Nagoya, Japan, 1995, pp. 1726–1731), it gives enough many voting to generate hypotheses. We also show that from the proposed invariant relationship, an invariant for the structure by two-view and an invariant for a structure proposed by Zhu et al. (Proceedings of the 12th International Conference on Robotics and Automation, Nagoya, Japan, 1995, pp. 1726–1731) can also be derived. Experiments using three-dimensional polyhedral objects are carried out to demonstrate the feasibility of our method for three-dimensional objects.

Classification of errors in locating a rigid body

This paper discusses the manner in which random Gaussian errors affect the determination of body segment kinematics. For the process of modelling rigid body (RB) motion, three types of kinematic errors, input, measured and theoretical, are identified. These correspond to errors in: the determination of three-dimensional observed points, the RB fit of those points, and the estimation of true RB positions, respectively. Of these, the theoretical error is most critical and most pivotal. Accuracy is provided when the theoretical error is minimised, yet only the measured error can be minimised by RB modelling algorithms. In computer simulations one may determine the effect that such manipulations have on theoretical error, yet in most experimental conditions this value may not even be calculated. Fortunately, computer simulations can be performed to determine the inter-relationships between the types of RB modelling errors. Such simulations can also be used to investigate the effects of RB shape. In this paper, Monte Carlo simulations were performed on three unit radius RBs; a triangle, a square and a tetrahedron. Although use of the triangle provided the lowest measured error, this also coincided with the greatest theoretical error. The use of redundant points was found to yield superior theoretical accuracies. A slight advantage was gained with use of the non-planar point arrangement on the tetrahedron, both the measured and theoretical errors were reduced. Finally, the superiority of RB modelling over individual point tracking was reflected in all of the results; between 33 and 50% of the input error was eliminated with the use of RB modelling.