Geometric feature enhanced line segment extraction from large-scale point clouds with hierarchical topological optimization

Hu, Zongtian; Chen, Chi; Wang, Jinhu; Wang, Zhiye; Ma, Ruiqi; Wu, Wenbo; Sun, Wenlu

doi:10.1016/j.jag.2022.102858

Cited by 8 publications

(3 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The challenge persists in merging data from both techniques to enhance monitoring accuracy, particularly when the monitoring precision of both techniques is within the centimetre range (within 10 cm). Various methods can be employed for this purpose (Chu et al, 2023;Hasheminasab et al, 2023;Hu et al, 2022;Jonassen et al, 2023;Li et al, 2018;Lv & Ren, 2015), including weighted averaging, height information fusion, feature matching and registration, geometric precision correction. This study, however, focuses on the stochastic nature of monitoring data (Gaussian distribution), further understanding the differences in deformation scales between the two monitoring data types, providing additional possibilities for data fusion.…”

Section: Discussionmentioning

confidence: 99%

Comparative analysis of surface deformation monitoring in a mining area based on UAV‐lidar and UAV photogrammetry

Zhan,

Zhang,

Wang

et al. 2024

The Photogrammetric Record

View full text Add to dashboard Cite

Unmanned aerial vehicle light detection and ranging (UAV‐lidar) and unmanned aerial vehicle (UAV) photogrammetry are currently commonly used surface monitoring technologies. Previous studies have used the two technologies interchangeably and ignored their correlation, or only compared them on a single product. However, there are few quantitative assessments of the differences between these two techniques in monitoring surface deformation and prediction of their application prospects. Therefore, the paper compared the differences between the digital elevation model (DEM) and subsidence basins obtained by the two techniques using Gaussian analysis. The results indicate that the surface DEMs obtained by both the techniques exhibit a high degree of similarity. The statistical analysis of the difference values in the z direction between the two DEMs follows a Gaussian distribution with a standard deviation of less than 0.36 m. When comparing the surface subsidence values monitored by the two techniques, it was found that UAV‐lidar was more sensitive to small‐scale deformation, with a difference range of 0.23–0.44 m compared to photogrammetry. The conclusion provides valuable information regarding the utilisation of multisource monitoring data.

show abstract

Section: Discussionmentioning

confidence: 99%

Comparative analysis of surface deformation monitoring in a mining area based on UAV‐lidar and UAV photogrammetry

Zhan,

Zhang,

Wang

et al. 2024

The Photogrammetric Record

View full text Add to dashboard Cite

show abstract

“…However, this required a labeled dataset, and it was not end-to-end. Hu et al [36] proposed a geometric feature enhanced line extraction method. Planes were first extracted by normal and curvature-based region growing and merging, and then line candidates were extracted by a 3D-2D projection, pretrained DexiNed [37] and MCMLSD [38] algorithm; finally, candidate lines were optimized by the hierarchical topology between contours and lines.…”

Section: Line Extractionmentioning

confidence: 99%

Simulation-Based Self-Supervised Line Extraction for LiDAR Odometry in Urban Road Scenes

Wang,

Zhou,

Dai

et al. 2023

Remote Sensing

View full text Add to dashboard Cite

LiDAR odometry is a fundamental task for high-precision map construction and real-time and accurate localization in autonomous driving. However, point clouds in urban road scenes acquired by vehicle-borne lasers are of large amounts, “near dense and far sparse” density, and contain different dynamic objects, leading to low efficiency and low accuracy of existing LiDAR odometry methods. To address the above issues, a simulation-based self-supervised line extraction in urban road scene is proposed, as a pre-processing for LiDAR odometry to reduce the amount of input and the interference from dynamic objects. A simulated dataset is first constructed according to the characteristics of point clouds in urban road scenes; and then, an EdgeConv-based network, named LO-LineNet, is used for pre-training; finally, a model transferring strategy is adopted to transfer the pre-trained model from a simulated dataset to real-world scenes without ground-truth labels. Experimental results on the KITTI Odometry Dataset and the Apollo SouthBay Dataset indicate that the proposed method can accurately extract reliable lines in urban road scenes in a self-supervised way, and the use of the extracted reliable lines as input for odometry can significantly improve its accuracy and efficiency in urban road scenes.

show abstract

“…-contact -non-contact -non-destructive -destructive. [1,6,7] This paper investigates non-destructive non-contact measurements using optical laser scanning. The data acquisition stage is then followed by the data processing stage which closely intersects with the third stage which is the fabrication of the component model.…”

Section: (A) Measurementmentioning

confidence: 99%

Data Preparing for Reverse Engineering

Kubátová,

Čížek,

Netrval

2023

10th Manufacturing Engineering Society International Conference (MESIC 2023)

View full text Add to dashboard Cite

Reverse engineering, or the creation of models from physical objects, is increasingly being used in many industries including manufacturing, automotive design, and computer animation. The presented work focuses on the issue of data preparation for reverse engineering. The aim of the thesis is to map the issues of working with the Zeiss LineScan optical sensor on a 3D coordinate measuring machine in the metrology laboratory of the Regional Institute of Technology at the University of West Bohemia. The aim of the work was to map the software setting options supplied by the sensor manufacturer. The sensor can obtain clouds of three types of points: RSL (Raw Scan Lines - measured points, directly recorded by the scanner, without further adjustments), GSL (Gridded Scan Lines - points reduced and grouped into a grid of a given step), and QSP (Qualified Surface points - compressed and grouped points into a given grid). The evaluation of the experiment revealed the ideal setting for the specified measurement conditions. Reference objects with nominal lengths of 100 mm and 500 mm were chosen. The data required for evaluating each sensor setting were obtained through a series of measurements of these objects. The output of the experiment is a comparison of each option for setting up the scanning method. GSL points with a step setting of 0.1 mm were evaluated as the most suitable setting for measuring objects with similar shapes. They showed stable deviations within the selected tolerance of ± 0.03 mm for all measurement variations. RSL and QSP points, on the other hand, showed less stability of deviations across the measurements.

show abstract

Geometric feature enhanced line segment extraction from large-scale point clouds with hierarchical topological optimization

Cited by 8 publications

References 36 publications

Comparative analysis of surface deformation monitoring in a mining area based on UAV‐lidar and UAV photogrammetry

Comparative analysis of surface deformation monitoring in a mining area based on UAV‐lidar and UAV photogrammetry

Simulation-Based Self-Supervised Line Extraction for LiDAR Odometry in Urban Road Scenes

Data Preparing for Reverse Engineering

Contact Info

Product

Resources

About