Accurate Point Cloud Research Articles

Semantic Segmentation of 3D Point Clouds in Outdoor Environments Based on Local Dual-Enhancement

Semantic segmentation of 3D point clouds in drivable areas is very important for unmanned vehicles. Due to the imbalance between the size of various outdoor scene objects and the sample size, the object boundaries are not clear, and small sample features cannot be extracted. As a result, the semantic segmentation accuracy of 3D point clouds in outdoor environment is not high. To solve these problems, we propose a local dual-enhancement network (LDE-Net) for semantic segmentation of 3D point clouds in outdoor environments for unmanned vehicles. The network is composed of local-global feature extraction modules, and a local feature aggregation classifier. The local-global feature extraction module captures both local and global features, which can improve the accuracy and robustness of semantic segmentation. The local feature aggregation classifier considers the feature information of neighboring points to ensure clarity of object boundaries and the high overall accuracy of semantic segmentation. Experimental results show that provides clearer boundaries between various objects, and has higher identification accuracy for small sample objects. The LDE-Net has good performance for semantic segmentation of 3D point clouds in outdoor environments.

Triple line-scan camera measurement for efficient and accurate narrow-space 3-D scanning with motion estimation.

Vision-based three-dimensional (3-D) shape measurement plays a crucial role in the inspection of vehicles and trains in the field of transportation. Consequently, the demand for more comprehensive narrow-space inspection has become an inevitable necessity and presents a great challenge to conventional vision methods. We propose the following efficient and accurate narrow-space 3-D scanning method based on triple line-scan cameras. First, the structure of coplanar dual line-scan cameras with a narrow reflector is proposed, making it easy for the optical path to enter a narrow space and obtain 3-D shape information. Efficient in-motion measurement with a large field of view is thereby achieved without the worries that often accompany narrow-space constraints. Secondly, an additional line-scan camera is attached to the coplanar dual cameras and creates a time-space constraint in point cloud stitching direction through the triple line-scan camera structure. With an efficient algorithm template including matching and motion estimation, accurate point cloud stitching is ensured. Lastly, point cloud texture mapping and light source optimization are realized. Our experimental results prove that the method realizes low-distortion in-motion reconstruction in narrow space despite motion variation.

AN EFFECTIVE APPROACH FOR POINT CLOUD DENOISING IN INTEGRATED SURVEYS

Abstract. Outliers and noise in point cloud data are unavoidable due to intrinsic and/or extrinsic survey factors. Significant errors may result from false geometry produced by a collection of anomalies, compounded by sparse structure, irregular densities, and lack of geometric cohesion typical of point clouds nature. Thus, filtering techniques on raw data are required to produce accurate point clouds suitable for further processing. This objective is pursued in the following study through a comparative analysis between two registered clouds, one obtained from TLS, used as reference dataset and the other – to be filtered – from SLAM system. Four steps make up the workflow: analysing the comparison models’ geometric attributes, specifically surface density and roughness; constructing statistical tolerance limits for the TLS cloud’s roughness distribution; cleaning the SLAM cloud; assessing the filtering outcomes. Our efforts to effectively remove and mitigate noise, while preserving the original detail features of the object surface, have been driven by the detailed articulation of point cloud denoising approaches that have been introduced in recent years. However, in this wide context, our goal is not to provide a review or to explore the details of the various methods; rather, we want to offer a simple yet efficient method for obtaining an integrated model with a uniform noise level. This can be especially useful when the data from the survey will later be used in source-based modelling.

Classification of Typical Static Objects in Road Scenes Based on LO-Net

Mobile LiDAR technology is a powerful tool that accurately captures spatial information about typical static objects in road scenes. However, the precise extraction and classification of these objects pose persistent technical challenges. In this paper, we employ a deep learning approach to tackle the point cloud classification problem. Despite the popularity of the PointNet++ network for direct point cloud processing, it encounters issues related to insufficient feature learning and low accuracy. To address these limitations, we introduce a novel layer-wise optimization network, LO-Net. Initially, LO-Net utilizes the set abstraction module from PointNet++ to extract initial local features. It further enhances these features through the edge convolution capabilities of GraphConv and optimizes them using the “Unite_module” for semantic enhancement. Finally, it employs a point cloud spatial pyramid joint pooling module, developed by the authors, for the multiscale pooling of final low-level local features. Combining three layers of local features, LO-Net sends them to the fully connected layer for accurate point cloud classification. Considering real-world scenarios, road scene data often consist of incomplete point cloud data due to factors such as occlusion. In contrast, models in public datasets are typically more complete but may not accurately reflect real-world conditions. To bridge this gap, we transformed road point cloud data collected by mobile LiDAR into a dataset suitable for network training. This dataset encompasses nine common road scene features; hence, we named it the Road9 dataset and conducted classification research based on this dataset. The experimental analysis demonstrates that the proposed algorithm model yielded favorable results on the public datasets ModelNet40, ModelNet10, and the Sydney Urban Objects Dataset, achieving accuracies of 91.2%, 94.2%, and 79.5%, respectively. On the custom road scene dataset, Road9, the algorithm model proposed in this paper demonstrated outstanding classification performance, achieving a classification accuracy of 98.5%.

Individual-Tree Segmentation from UAV–LiDAR Data Using a Region-Growing Segmentation and Supervoxel-Weighted Fuzzy Clustering Approach

Accurate individual-tree segmentation is essential for precision forestry. In previous studies, the canopy height model-based method was convenient to process, but its performance was limited owing to the loss of 3D information, and point-based methods usually had high computational costs. Although some hybrid methods have been proposed to solve the above problems, most canopy height model-based methods are used to detect subdominant trees in one coarse crown and disregard the over-segmentation and accurate segmentation of the crown boundaries. This study introduces a combined approach, tested for the first time, for treetop detection and tree crown segmentation using UAV–LiDAR data. First, a multiscale adaptive local maximum filter was proposed to detect treetops accurately, and a Dalponte region-growing method was introduced to achieve crown delineation. Then, based on the coarse-crown result, the mean-shift voxelization and supervoxel-weighted fuzzy c-means clustering method were used to identify the constrained region of each tree. Finally, accurate individual-tree point clouds were obtained. The experiment was conducted using a synthetic uncrewed aerial vehicle (UAV)–LiDAR dataset with 21 approximately 30 × 30 m plots and an actual UAV–LiDAR dataset. To evaluate the performance of the proposed method, the accuracy of the remotely sensed biophysical observations and retrieval frameworks was determined using the tree location, tree height, and crown area. The results show that the proposed method was efficient and outperformed other existing methods.

BIMBot for Autonomous Laser Scanning in Built Environments

Accurate and complete 3D point clouds are essential in creating as-built building information modeling (BIM) models, although there are challenges in automating the process for 3D point cloud creation in complex environments. In this paper, an autonomous scanning system named BIMBot is introduced, which integrates advanced light detection and ranging (LiDAR) technology with robotics to create 3D point clouds. Using our specially developed algorithmic pipeline for point cloud processing, iterative registration refinement, and next best view (NBV) calculation, this system facilitates an efficient, accurate, and fully autonomous scanning process. The BIMBot’s performance was validated using a case study in a campus laboratory, featuring complex structural and mechanical, electrical, and plumbing (MEP) elements. The experimental results showed that the autonomous scanning system produced 3D point cloud mappings in fewer scans than the manual method while maintaining comparable detail and accuracy, demonstrating its potential for wider application in complex built environments.

Unlocking the full potential of reality capture for construction project cost monitoring: a mixed review

Purpose Digital data acquisition is crucial for operations in the digital transformation era. Reality capture (RC) has made an immeasurable contribution to various fields, especially in the built environment. This paper aims to review RC applications, potentials, limitations and the extent to which RC can be adopted for cost monitoring of construction projects. Design/methodology/approach A mixed-method approach, using Bibliometric analysis and the PRISMA framework, was used to review and analyse 112 peer-reviewed journal articles from the Scopus and Web of Science databases. Findings The study reveals RC has been applied in various areas in the built environment, but health and safety, cost and labour productivity monitoring have received little or no attention. It is proposed that RC can significantly support cost monitoring owing to its ability to acquire accurate and quick digital as-built 3D point cloud data, which contains rich measurement points for the valuation of work done. Research limitations/implications The study’s conclusions are based only on the Scopus and Web of Science data sets. Only English language documents were approved, whereas others may be in other languages. The research is a non-validation of findings using empirical data to confirm the data obtained from RC literature. Practical implications This paper highlights the importance of RC for cost monitoring in construction projects, filling knowledge gaps and enhancing project outcomes. Social implications The implementation of RC in the era of the digital revolution has the potential to improve project delivery around the world today. Every project’s success is largely determined by the availability of precise and detailed digital data. RC applications have pushed for more sustainable design, construction and operations in the built environment. Originality/value The study has given research trends on the extent of RC applications, potentials, limitations and future directions.

Structure from Motion-Based Mapping for Autonomous Driving: Practice and Experience

Accurate and up-to-date 3D maps, often represented as point clouds, are crucial for autonomous vehicles. Crowd-sourcing has emerged as a low-cost and scalable approach for collecting mapping data utilizing widely available dashcams and other sensing devices. However, it is still a non-trivial task to utilize crowdsourced data, such as dashcam images and video, to efficiently create or update high-quality point clouds using technologies like Structure from Motion (SfM). This study assesses and compares different image matching options available in open-source SfM software, analyzing their applicability and limitations for mapping urban scenes in different practical scenarios. Furthermore, the study analyzes the impact of various camera setups (i.e., the number of cameras and their placement) and weather conditions on the quality of the generated 3D point clouds in terms of completeness and accuracy. Based on these analyses, our study provides guidelines for creating more accurate point clouds.

A Transformer-based multi-modal fusion network for 6D pose estimation

6D pose estimation has garnered significant attention and research. RGB images and point clouds converted from RGB-D images provide complementary color and geometry information, making them the mainstream data sources for object 6D pose estimation. However, due to the fact that RGB image and point cloud belong to different dimensional spaces and have different distribution characteristics, the fusion of these two complementary data sources remains a key technical challenge for 6D pose estimation. In contrast to prior approaches that simply concatenate separately processed RGB images and point clouds, this work introduces a Transformer-based multi-modal fusion network to address this challenge. More precisely, We build a Transformer architecture based pixel-wise feature extraction to optimize feature extraction from RGB images and point clouds. Subsequently, we investigate various multi-modal feature fusion methods to process these features, enabling deeper fusion of complementary data. Additionally, during the experimental phase, we design a 6D pose estimation network based on depth prediction to assess the impact of point cloud accuracy on the multi-modal fusion module. Finally, the proposed method is verified on four datasets: LineMOD, Occlusion Linemod, MP6D and YCB-Video. Experimental results show that the proposed method outperforms similar methods on these datasets.

The influence of the incident angles variation on the accuracy of TLS point cloud based on surface reflectivity and roughness

The influence of the incident angles variation on the accuracy of TLS point cloud based on surface reflectivity and roughness

Point Cloud Registration in Laparoscopic Liver Surgery Using Keypoint Correspondence Registration Network.

Laparoscopic liver surgery is a newly developed minimally invasive technique and represents an inevitable trend in the future development of surgical methods. By using augmented reality (AR) technology to overlay preoperative CT models with intraoperative laparoscopic videos, surgeons can accurately locate blood vessels and tumors, significantly enhancing the safety and precision of surgeries. Point cloud registration technology is key to achieving this effect. However, there are two major challenges in registering the CT model with the point cloud surface reconstructed from intraoperative laparoscopy. First, the surface features of the organ are not prominent. Second, due to the limited field of view of the laparoscope, the reconstructed surface typically represents only a very small portion of the entire organ. To address these issues, this paper proposes the keypoint correspondence registration network (KCR-Net). This network first uses the neighborhood feature fusion module (NFFM) to aggregate and interact features from different regions and structures within a pair of point clouds to obtain comprehensive feature representations. Then, through correspondence generation, it directly generates keypoints and their corresponding weights, with keypoints located in the common structures of the point clouds to be registered, and corresponding weights learned automatically by the network. This approach enables accurate point cloud registration even under conditions of extremely low overlap. Experiments conducted on the ModelNet40, 3Dircadb, DePoLL demonstrate that our method achieves excellent registration accuracy and is capable of meeting the requirements of real-world scenarios.

High-Resolution Point Cloud Registration Method for Three-Dimensional Piping Measurements

In this study, we propose a method for generating highly accurate high-density point clouds of piping facilities using an unmanned aerial vehicle (UAV) laser scanner and a handheld laser scanner. The point cloud for each scanline measured by the UAV scanner is repositioned on the piping axis, and the handheld scanner’s 3D point cloud is subsequently registered so that the center axis of the piping coincides with the UAV point cloud as a reference. The method proposed in this study was used to accurately reconstruct linear piping measured in high winds, which can easily deteriorate measurement accuracy. Whereas the conventional method incurred a deviation of 44.3 mm between the predicted and true values at altitudes of 15 m, the proposed method reduced this deviation to 19.4 mm. An application of the registration method demonstrated that the combined use of the two laser scanners enabled the creation of a high-density point cloud.

DEVELOPING COMPLETE URBAN DIGITAL TWINS IN BUSY ENVIRONMENTS: A FRAMEWORK FOR FACILITATING 3D MODEL GENERATION FROM MULTI-SOURCE POINT CLOUD DATA

Abstract. The proliferation of affordable LiDAR technology and photogrammetry sensors has revolutionized 3D data acquisition in built environments, enabling comprehensive data capture from citywide scales to interior structures. This data can be transformed into digital twins, providing valuable resources for city planners, architects, engineers, and decision-makers. However, current studies often overlook the limitations of real-world point cloud datasets derived from LiDAR systems, which are voluminous, noisy, incomplete, and lacking information, which hinders monitoring, interpretation, and automated analysis. To address these challenges, methods are required to prepare point cloud data, ensuring accurate and reliable 3D representations. This research proposes a detailed framework for point cloud data preparation in busy urban environments. It includes precise algorithms, software, and parameter guidelines, allowing for the creation of comprehensive point cloud datasets. The framework has been successfully implemented on datasets acquired in Toronto, converting point cloud data from various platforms and parameters into an integrated dataset. Results demonstrate the framework's effectiveness in producing accurate and complete point cloud datasets for applications such as classification, information extraction, 3D model generation, and smart cities' monitoring and management.

Defect segmentation with local embedding in industrial 3D point clouds based on transformer

Three-dimensional (3D) defect detection provides an effective method for improving industrial production efficiency. However, the 3D dataset is scarce, which is valuable for the industrial production field. This study proposes a new approach for detecting defect point clouds, which can provide an end-to-end 3D defect detection model. A self-attention mechanism is used to enrich the semantic relationships between local neighborhood features and global features based on the connection between them. Through adding multi-channel features, the rich structural features of the target point cloud are obtained, and the defect areas are accurately segmented to finally complete the 3D point cloud defect detection task. Furthermore, the multi-feature fusion in the model makes the segmented defect regions closer to the ground truth. Our method outperforms four state-of-the-art point cloud segmentation methods in terms of both segmentation region accuracy and defect detection point cloud accuracy. In the field of 3D defect detection, it provides an effective method to detect 3D information of industrial products.

Enhancing leaf area index and biomass estimation in maize with feature augmentation from unmanned aerial vehicle-based nadir and cross-circling oblique photography

Rapid and accurate estimation of plant phenotypes plays a vital role in effective breeding and management of maize crops. Unmanned aerial vehicle (UAV) platforms are emerging as a valuable tool in the assessment of crop phenotypes, offering a promising avenue to enhance the accuracy and efficiency of phenotypic analysis. This study executed two UAV photography methods to acquire aerial images for the estimation of leaf area index (LAI) and above-ground biomass (AGB) in summer maize. Nadir photography was implemented to obtain data from three sources including multi-spectral (MS), RGB, and thermal infrared (TIR) sensors. Additionally, RGB imaging-based cross-circling oblique (CCO) photography was implemented to obtain accurate 3D point cloud data. The nadir photography data was processed to generate orthomosaic image and extract features including vegetation index, canopy cover, canopy height, canopy temperature, and textural information. Features including canopy occupation volume, plant area index, canopy cover, and canopy height were extracted from the CCO photography-derived point cloud data. Furthermore, a data augmentation method, called linear regression-based feature augmentation (LRFA), was proposed for augmenting features extracted from nadir and CCO photography. The results showed that the introduction of CCO photography improved the accuracy of LAI and AGB estimation during the big trumpet stage and LAI estimation during the milk stage compared to the integrated multi-source nadir photography data. Notably, the LRFA derived features outperformed raw features in the majority of modeling scenarios, with the random forest achieved an average accuracy (R2) improvement of 6.1% for LAI estimation and 3.7% for AGB estimation. This study highlights the significance of combining different photography technologies and feature augmentation method for estimating maize phenotypes, providing novel opportunities for crop growth monitoring in modern agriculture.

Reverse design of involute helical gear pairs considering machining errors based on IMAGEWARE and HYPERMESH

Gears inevitably contain errors during the manufacturing and installation process, which can increase the vibration and noise of the gear system. The finite element model with error can accurately solve the variation of vibration and noise and meshing characteristics of such gear systems. This paper develops a combined analytical and simulation method for involute helical gears when machining errors are included. First, the tooth profile accurate point clouds of the involute gear are derived by cutting the tooth blank with the rack and the coordinate transformation function. Subsequently, the gear is reconstructed by the IMAGEWARE and UG. With HYPERMESH's powerful node editing capabilities, sinusoidal profile errors can be assigned to gears. Also, the intersection of the error gear with the tooth blank is used to find the apex, which has an essential effect on noise. Finally, the difference in static transmission errors of gears with and without errors is analyzed by ABAQUS. The results show that gears containing errors increase the peak-to-peak values of transmission errors, i.e., increasing noise. The method proposed in this paper can provide a basis for modeling and analysis of meshing characteristics of special gears that cannot be directly parameterized or gears with various errors.

Line feature based self-calibration method for dual-axis scanning LiDAR system

LiDAR calibration is an essential step for acquiring accurate three-dimensional point-cloud. To enhance the convenience of calibration of dual-axis scanning LiDAR, a line feature based self-calibration method is proposed, which corrects point-cloud error by establishing observation model and estimating model parameters. The model parameters are estimated by minimizing the square error of line feature points, referring as straight edges extracted from point-cloud. The experimental results verify the effectiveness of the proposed method on improving point-cloud accuracy, the horizontal and vertical angular errors are reduced to 0.10° and 0.08° after calibration with reductions of 64% and 68%, respectively. The analysis on parameter correlation finds that, parameters can be divided into groups according to the introduced point-cloud error: dimension error, trapezoid and parallelogram shape distortion. A de-correlation strategy combining inter-group and stepwise intra-group de-correlation thus can be implemented, which can be applied for point-cloud quality enhancement in in-situ environmental perception scenes.

A Method for Point Cloud Accuracy Analysis Based on Intensity Information.

Three-dimensional laser scanning has emerged as a prevalent measurement method in numerous high-precision applications, and the precision of the obtained data is closely related to the intensity information. Comprehending the association between intensity and point cloud accuracy facilitates scanner performance assessment, optimization of data acquisition strategies, and evaluation of point cloud precision, thereby ensuring data reliability for high-precision applications. In this study, we investigated the correlation between point cloud accuracy and two distinct types of intensity information. In addition, we presented methods for assessing point cloud accuracy using these two forms of intensity information, along with their applicable scopes. By examining the percentage intensity, we analyzed the reflectance properties of the scanned object's surface employing the Lambertian model. Our findings indicate that the Lambertian circle fitting radius is inversely correlated with the scanner's ranging error at a constant scanning distance. Experimental outcomes substantiate that modifying the surface characteristics of the object enables the attainment of higher-precision point cloud data. By constructing a model associating the raw reflectance intensity with ranging errors, we developed a single-point error ellipsoid model to assess the accuracy of individual points within the point cloud. The experiments revealed that the ranging error model based on the raw intensity is solely applicable to point cloud data unaffected by specular reflectance properties. Moreover, the devised single-point error ellipsoid model accurately evaluates the measurement error of individual points. Both analytical methods can be utilized to evaluate the performance of the scanner as well as the accuracy of the acquired point cloud data, providing reliable data support for various high-precision applications.

RoCNet++: Triangle-based descriptor for accurate and robust point cloud registration

This paper introduces RoCNet++, a point cloud registration method with two main contributions, one concerning the design of a robust descriptor and another concerning the estimation of the rigid transformation. First, to robustly capture the local geometric properties of the surface, i.e., each point is characterized by all the triangles formed by itself and its nearest neighbours in the 3D point cloud. The idea is to assist the learning of the descriptor by introducing a priori information about interesting geometric properties such as the invariance of triangle angles under rigid transformations. This local triangle-based descriptor is integrated into the recently developed RoCNet architecture for estimating the correspondences between source and target point clouds. We then introduce the Farthest Sampling-guided Registration (FSR), which relies on successive farthest point samplings to estimate the global rigid transformation between 3D point clouds. The new proposed architecture RoCNet++ has been evaluated in different configurations: clean, noisy and partial data on both synthetic and real databases such as ModelNet40, KITTI, and 3DMatch. RoCNet++ shows improved performances on these benchmark datasets in favourable and unfavourable conditions. Furthermore, both the local triangle-based descriptor and the Farthest Sampling-guided Registration (FSR) can be used in other registration algorithms.

Coaxiality prediction for aeroengines precision assembly based on geometric distribution error model and point cloud deep learning

Assembly accuracy of aeroengines influences operation performance and service life. The coaxiality of the aeroengine is the main index of assembly accuracy and is also a core index to represent assembly quality. However, direct measurement of coaxiality is a difficult technical problem due to the sealed structure of the aeroengine casing system. A coaxiality prediction method is proposed to obtain coaxiality and assist assembly by geometric distribution error modeling and point cloud deep learning. The prediction process consists of three steps. In the beginning, the geometric distribution error model is established to construct the accurate dense point cloud of aeroengine part surfaces by the non-uniform rational B-splines (NURBS) method based on the coordinate measuring machine collecting information. Then, the mapping between the dense point cloud and coaxiality is established to obtain an assembly dataset by the virtual assembly. Finally, the dataset is fed to a new point cloud deep learning backbone, Self-channel cross attention point network, and realizes end-to-end coaxiality prediction based on the aeroengine surface point cloud. The geometric distribution error model is tested on the aeroengine simulated parts with 0.001 mm accuracy. The prediction method is verified on the aeroengine simulated parts and compared with other point cloud deep learning baselines. The method proposed in this paper realizes 93.17% prediction accuracy with 0.01 mm coaxiality precision which is a high performance and meets the requirements of industrial measurement. This paper provides an effective coaxiality prediction model for the aeroengine casing system, to improve the accuracy and efficiency of the aeroengine assembly.