Optimization Algorithm for Point Cloud Quality Enhancement Based on Statistical Filtering

Point clouds are widely used to construct models of workpieces using 3D scanners, especially where high-quality robotic and automatic 3D scanning is required in industries and manufacturing. In recent years, Point Cloud Quality Assessment (PCQA) has garnered increasing attention as it provides quality scores for entire point clouds, addressing issues such as downsampling and compression distortions. However, current PCQA methods cannot provide specific and local quality scores, which are necessary to facilitate rescanning and recompletion in 3D robotic scanning. Additionally, 3D data produced by robotic view-planning algorithms are usually considered the final result, where PCQA is typically not involved. In this paper, we bridge the gap between PCQA methods and practical robotic 3D scanning. We propose a no-reference PCQA method that recognizes sparse regions during 3D scanning, providing both local and overall quality scores. Unlike traditional methods that primarily consider density as a key metric, our method assumes that an expected 3D scan will have a uniformly distributed point cloud on surfaces. We analyze the quality of points by using geometric information from surfaces fitted to these points, which are mapped to a 2D distribution based on specified distances and angles. We conducted experiments on various datasets, including both synthetic and public datasets, to evaluate the accuracy and robustness of our method. The results show that our method can represent the quality on surfaces more accurately and robustly than density calculation methods. Additionally, it outperforms most existing PCQA methods in scenarios of downsampling, which is a common challenge in high-quality 3D scanning applications. The performance of our quality enhancement experiments on practical 3D scanning, conducted towards the end of our study, demonstrates significant potential for real-world applications. The related code is released at https://github.com/leihui6/PCQA.

Affordable prices for 3D laser range finders and mature software solutions for registering multiple point clouds in a common coordinate system paved the way for new areas of application for 3D point clouds. Nowadays we see 3D laser scanners being used not only by digital surveying experts but also by law enforcement officials, construction workers or archaeologists. Whether the purpose is digitizing factory production lines, preserving historic sites as digital heritage or recording environments for gaming or virtual reality applications -- it is hard to imagine a scenario in which the final point cloud must also contain the points of objects like factory workers, pedestrians, cars or flocks of birds. For most post-processing tasks, moving objects are undesirable not least because moving objects will appear in scans multiple times or are distorted due to their motion relative to the scanner rotation. The main contributions of this work are two postprocessing steps for already registered 3D point clouds. The first method is a new change detection approach based on a voxel grid which allows partitioning the input points into static and dynamic points using explicit change detection and subsequently remove the latter for a point cloud. The second method uses this cleaned point cloud as input for detecting collisions between points of the environment point cloud and a point cloud of a model that is moved through the scene. Our approach on explicit change detection is compared to the state of the art using multiple datasets including the popular KITTI dataset. We show how our solution achieves similar or better F1-scores than an existing solution while at the same time being faster. To detect collisions we do not produce a mesh but approximate the raw point cloud data by spheres or cylindrical volumes. We show how our data structures allow efficient nearest neighbor queries that make our CPU-only approach comparable to a massively-parallel algorithm running on a GPU. The utilized algorithms and data structures are discussed in detail. All our software is freely available for download under the terms of the GNU General Public license. Most of the datasets used in this thesis are freely available as well. We provide shell scripts that allow one to directly reproduce the quantitative results shown in this thesis for easy verification of our findings.

Accurately predicting point cloud quality plays an important role in human vision tasks. This paper presents an effective and robust objective point cloud quality assessment model called elastic potential energy similarity (EPES). Motivated by the knowledge on point cloud distortion, EPES first expresses a point cloud as a collection of spatially scattered points. A set of origins are then deployed and the scattered points are assumed to be connected to the nearest origin using <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">springs</i> . Imposing external forces can move the scattered points to specific locations such that the resulting point clouds would exhibit desired characteristics. At the same time, this process will store elastic potential energies in the springs. Therefore, through comparing the elastic potential energies kept in the springs of the reference and distorted point clouds, we are able to quantify the influence of distortion on the point cloud quality. The proposed quality assessment model is evaluated on three fairly large databases, SJTU-PCQA, CPCQA, and LSPCQA. Experimental results show that EPES is superior to the state-of-the-art metrics. Ablation studies demonstrate that the developed EPES is robust to variations in the model parameter settings.

Statistical process control (SPC) methods have been extensively applied to monitor the quality performance of manufacturing processes to quickly detect and correct out-of-control conditions. As sensor and measurement technologies advance, there is a continual need to adapt and refine SPC methods to effectively and efficiently use these new data-sets. One of the most state-of-the-art dimensional measurement technologies currently being implemented in industry is the 3D laser scanner, which rapidly provides millions of data points to represent an entire manufactured part's surface. Consequently, this data has a great potential to detect unexpected faults, i.e., faults that are not captured by measuring a small number of predefined dimensions. However, in order for this potential to be realized, SPC methods capable of handling these large data-sets need to be developed. This paper presents an approach to performing SPC using point clouds obtained through a 3D laser scanner. The proposed approach transforms high-dimensional point clouds into linear profiles through the use of Q---Q plots, which can be monitored by well established profile monitoring techniques. In this paper point clouds are simulated to determine the performance of the proposed approach under varying fault scenarios. In addition, experimental studies were performed to determine the effectiveness of the proposed approach using actual point cloud data. The results of these experiments show that the proposed approach can significantly improve the monitoring capabilities for manufacturing parts that are characterized by complex surface geometries.

In recent years, a 3D laser scanner has increasing attention as one of the modern technologies to digitizing the 3D object surface in manner to obtain a mathematical representation of object's and surface reconstruction. Due to limitations of a 3D laser scanners, the row point cloud, which are acquired from these techniques, included some undesired information such as noise points and associated huge number of points in point cloud. In the present paper, a new proposed point cloud simplification algorithm for scanned object using 3D laser scanner (Matter and Form) has been adopted in manner to extract the necessary geometric features which represented by a 3D surface curvature. This algorithm based on instantaneous calculation of 3D Gaussian curvature for each data point. The curvature value indicates both quality of the topical calculations and a noise isolating and detection in the point cloud set. A MATLAB environment has been adopted to construct a proposed point cloud simplification algorithm program, this program has been proved using a practically case studies. The results proved the effectiveness of the proposed algorithm in local calculation of 3D curvature and detection and isolate the noise points in point cloud where the percent of data which was ignored as noisy data point were (49.29%) and (57.30%), of total data point number after applying the algorithm for the first and second proposed case studies respectively.

This passage puts forward a kind of simple and convenient three-dimensional laser measurement technology for a small volume and complex shape space data acquisition of the object. This technology mainly uses the camera self calibration method for construction of three-dimensional model algorithm and creates the measurement environment of the measuring technique in machinery and hardware camp. The result of the experiment shows that the desktop 3D laser scanning measuring instrument is convenient, and can acquire spatial location ideal data with low cost.

When 3D laser scanning (LIDAR) is used for navigation of autonomous vehicles operated on unstructured terrain, it is necessary to register the acquired point cloud and accurately perform point cloud reconstruction of the terrain in time. This paper proposes a novel registration method to deal with uneven-density and high-noise of unstructured terrain point clouds. It has two steps of operation, namely initial registration and accurate registration. Multisensor data is firstly used for initial registration. An improved Iterative Closest Point (ICP) algorithm is then deployed for accurate registration. This algorithm extracts key points and builds feature descriptors based on the neighborhood normal vector, point cloud density and curvature. An adaptive threshold is introduced to accelerate iterative convergence. Experimental results are given to show that our two-step registration method can effectively solve the uneven-density and high-noise problem in registration of unstructured terrain point clouds, thereby improving the accuracy of terrain point cloud reconstruction.

The density of acquired data by 3D laser scanner is huge and is not convenient for curved face reconstruction, model reconstruction, data storage, data display and transmission.Firstly, the conventional point cloud data classification and simplification method is introduced.Secondly, the angle deviation error and chord deviation error simplification method is researched based on the analysis of characteristic of 3D laser scan point cloud data.Finally, the angle--chord comprehensive simplification method is prompted.The example shows that this method could simplify the scan line point cloud data got by 3D laser scanner directly and effectively.This method could retain the point cloud data which reflects the surface character of object with higher simplification efficiency.

Nowadays, point clouds are frequently gathered by 3D scanners such as Lidar and Kinect, which produces thousands of point cloud models. Point cloud processing is vital to 3D vision, especially 3D object recognition, positioning, and navigation technology. Addressing uneven data density caused by coordinate frame transformations and the inherent problem of insufficient context connection in point clouds, the DiM-PCNet (Multi-scale and Multi-level Point Clouds Classification Network, and the Di is a prefix to represent double M in Multi-scale and Multi-level) is proposed in this paper. DiM-PCNet is provided for object classification with multi-scale and multi-level features. We encode the point cloud in multi-scale and fully fuse the features with the raw point cloud for keeping the context relationship. In DiM-PCNet, we sample the point clouds from eight parts for multi-scales feature extraction. The multi-scales features are fully fused by multi-level pyramid models. The multi-scale and multi-level strategies are applied in DiM-PCNet, in which the abundant and important features of point clouds are extracted and utilized in the 3D object classification. It is worth noting that the DiM-PCNet feature block can be embedded into the segmentation net, where the accuracy achieved is 87.1%. We conducted experiments on ShapeNet and ModelNet40 and the experimental results show that DiM-PCNet achieves state-of-the-art performance in 3D object classification. The experiment shows competitive performance on robustness and segmentation tasks.

In industrial production, the disordered grasping operation of the robotic arm is mostly for grasping a single type of workpiece. Effective grasping is not easy when multiple overlapping workpieces are mixed together. The mutual occlusion between workpieces causes the loss of geometric shape information, which makes it difficult to obtain the precise grasping pose of each workpiece. In this paper, a 3D laser scanner is used to acquire the point cloud features of the workpiece. At first, the point clouds are filtered and segmented, and then they are input into the RS-CNN network for recognition and classification. According to the classification results, different models are used to register the point clouds in the scene. At last, the final pose of the workpiece to be grasped is obtained, which realizes the disorderly grasping of various workpieces.

Point clouds are a crucial data type for mmWave radar and are widely used in sensing applications such as human tracking and activity recognition. However, for indoor human sensing, the point clouds obtained by mmWave radar are often sparse. Previous studies attribute the sparsity to the limited sensing capabilities of mmWave radar, underestimating the impact of CFAR---a key algorithmic component of mmWave systems---on point cloud quality. Through empirical studies, we find that the spatial-based CFAR widely used in existing works suffers from a severe energy masking issue. This is because these algorithms work well when the target is far away enough to be approximated as a point. In short-range indoor sensing, the human body can not be considered as a point but an extended target, causing the spatial-based CFAR to calculate the noise power wrongly and accordingly a miss-generation of the point cloud. To fundamentally solve the problem, this paper proposes a temporal-based CFAR named ETCM-CFAR. We address multiple issues such as lacking initial noise power and the absence of a closed-form threshold solution to make the proposed algorithm work. Based on ETCM-CFAR, this paper proposes a point cloud generation system named mmPC. mmPC is implemented on three different types of commercial-off-the-shelf mmWave radars and extensive experiments demonstrate that mmPC significantly improves point cloud quality, increasing the number of cloud points by 148.6% compared to the state-of-the-art systems. Two representative sensing applications, i.e., fitness activity recognition and human-pet classification are further employed to demonstrate the effectiveness of mmPC on sensing.

As buildings have almost come to a saturation point in most developed countries, the management and maintenance of existing buildings have become the major problem of the field. Building Information Modeling (BIM) is the underlying technology to solve this problem. It is a 3D semantic representation of building construction and facilities that contributes to not only the design phase but also the construction and maintenance phases, such as life-cycle management and building energy performance measurement. This study aims at the processes of creating as-built BIM models, which are constructed after the design phase. Point cloud, a set of points in 3D space, is an intermediate product of as-built BIM models that is often acquired by 3D laser scanning and photogrammetry. A raw point cloud typically requires further procedures, e.g. registration, segmentation, classification, etc. In terms of segmentation and classification, machine learning methodologies are trending due to the enhanced speed of computation. However, supervised machine learning methodologies require labelling the training point clouds in advance, which is time-consuming and often leads to inevitable errors. And due to the complexity and uncertainty of real-world environments, the attributes of one point vary from the attributes of others. These situations make it difficult to analyze how one single attribute contributes to the result of segmentation and classification. This study developed a method of producing point clouds from a fast-generating 3D virtual indoor environment using procedural modeling. This research focused on two attributes of simulated point clouds, point density and the level of random errors. According to Silverman (1986), point density is associated with the point features around each output raster cell. The number of points within a neighborhood divided the area of the neighborhood is the point density. However, in this study, there was a little different. The point density was defined as the number of points on a surface divided by the surface area. And the unit is points per square meters (pts/m2). This research compared the performances of a machine learning segmentation and classification algorithm on ten different point cloud datasets. The mean loss and accuracy of segmentation and classification were analyzed and evaluated to show how the point density and level of random errors affect the performance of the segmentation and classification models. Moreover, the real-world point cloud data were used as additional data to evaluate the applicability of produced models.

Point clouds obtained from 3D scanners are often noisy and cannot be directly used for subsequent high-level tasks. In this article, we propose a novel point cloud optimization method capable of denoising and homogenizing point clouds. Our idea is based on the assumption that the noise is generally much smaller than the effective signal. We perform noise perturbation on the noisy point cloud to get a new noisy point cloud, called self-variation point cloud. The noisy point cloud and self-variation point cloud have different noise distribution, but the same point cloud distribution. We compute the potential commonality between two noisy point clouds to obtain a clean point cloud. To implement our idea, we propose a Self-Variation Capture Network (SVCNet). We perturb the point cloud features in the latent space to obtain self-variation feature vectors, and capture the commonality between two noisy feature vectors through the feature aggregation and averaging. In addition, an edge constraint module is introduced to suppress low-pass effects during denoising. Our denoising method does not take into account the noise characteristics, and can filter the drift noise located on the underlying surface, resulting in a uniform distribution of the generated point cloud. The experimental results show that our algorithm outperforms the current state-of-the-art algorithms, especially in generating more uniform point clouds. In addition, extended experiments demonstrate the potential of our algorithm for point clouds upsampling.

Afterthe construction of subway main structure, in order to realize route adjustmentof alignment and gradient, it needs to survey the cross-section of subwaytunnel. Compared with conventional measuring methods, 3D laser scanning has thecharacteristics of non-contact measurement and can collect space 3D point clouddata with high density, this can improve the working efficiency for the subwaycross-section surveying. Based on the Leica Scanstation 2 scanners this paperanalyzed the 3D laser scanning point cloud data collection procedures and dataprocessing, expounded the subway cross-section surveying method based on pointcloud data; analyzed the feasibility of 3D laser scanning technology in theapplication of tunnel cross-section surveying based on the field validationdata. The results show that the cross-section measured by this method can meetthe technical requirements of route adjustment of alignment and gradient.

In recent years, three-dimensional (3D) terrestrial laser scanning technologies with higher precision and higher capability are developing rapidly. The growing maturity of laser scanning has gradually approached the required precision as those have been provided by traditional structural monitoring technologies. Together with widely available fast computation for massive point cloud data processing, 3D laser scanning can serve as an efficient structural monitoring alternative for civil engineering communities. Currently most research efforts have focused on integrating/calculating the measured multi-station point cloud data, as well as modeling/establishing the 3D meshes of the scanned objects. Very little attention has been spent on extracting the information related to health conditions and mechanical states of structures. In this study, an automated numerical approach that integrates various existing algorithms for geometric identification and damage detection of structural elements were established. Specifically, adaptive meshes were employed for classifying the point cloud data of the structural elements, and detecting the associated damages from the calculated eigenvalues in each area of the structural element. Furthermore, kd-tree was used to enhance the searching efficiency of plane fitting which were later used for identifying the boundaries of structural elements. The results of geometric identification were compared with M3C2 algorithm provided by CloudCompare, as well as validated by LVDT measurements of full-scale reinforced concrete beams tested in laboratory. It shows that 3D laser scanning, through the established processing approaches of the point cloud data, can offer a rapid, nondestructive, remote, and accurate solution for geometric identification and damage detection of structural elements.