Plane segmentation and fitting method of point clouds based on improved density clustering algorithm for laser radar

2019 ◽  
Vol 96 ◽  
pp. 133-140 ◽  
Author(s):  
Xiaobin Xu ◽  
Minzhou Luo ◽  
Zhiying Tan ◽  
Min Zhang ◽  
Hao Yang
Keyword(s):  
Clustering Algorithm ◽  
Point Clouds ◽  
Laser Radar ◽  
Fitting Method ◽  
Density Clustering

scholarly journals Higher-Order Conditional Random Fields-Based 3D Semantic Labeling of Airborne Laser-Scanning Point Clouds

2019 ◽  
Vol 11 (10) ◽  
pp. 1248 ◽  
Author(s):  
Yong Li ◽  
Dong Chen ◽  
Xiance Du ◽  
Shaobo Xia ◽  
Yuliang Wang ◽  
...  
Keyword(s):  
Random Fields ◽  
Laser Scanning ◽  
Clustering Algorithm ◽  
Conditional Random Fields ◽  
Point Clouds ◽  
Higher Order ◽  
Fine Grained ◽  
Airborne Laser ◽  
Semantic Labeling ◽  
Density Clustering

This paper presents a novel framework to achieve 3D semantic labeling of objects (e.g., trees, buildings, and vehicles) from airborne laser-scanning point clouds. To this end, we propose a framework which consists of hierarchical clustering and higher-order conditional random fields (CRF) labeling. In the hierarchical clustering, the raw point clouds are over-segmented into a set of fine-grained clusters by integrating the point density clustering and the classic K-means clustering algorithm, followed by the proposed probability density clustering algorithm. Through this process, we not only obtain a more uniform size and more homogeneous clusters with semantic consistency, but the topological relationships of the cluster’s neighborhood are implicitly maintained by turning the problem of topology maintenance into a clustering problem based on the proposed probability density clustering algorithm. Subsequently, the fine-grained clusters and their topological context are fed into the CRF labeling step, from which the fine-grained cluster’s semantic labels are learned and determined by solving a multi-label energy minimization formulation, which simultaneously considers the unary, pairwise, and higher-order potentials. Our experiments of classifying urban and residential scenes demonstrate that the proposed approach reaches 88.5% and 86.1% of “m F 1 ” estimated by averaging all classes of the F 1 -scores. We prove that the proposed method outperforms five other state-of-the-art methods. In addition, we demonstrate the effectiveness of the proposed energy terms by using an “ablation study” strategy.

scholarly journals Canopy Parameter Estimation of Citrus grandis var. Longanyou Based on LiDAR 3D Point Clouds

2021 ◽  
Vol 13 (9) ◽  
pp. 1859
Author(s):  
Xiangyang Liu ◽  
Yaxiong Wang ◽  
Feng Kang ◽  
Yang Yue ◽  
Yongjun Zheng
Keyword(s):  
Convex Hull ◽  
Clustering Algorithm ◽  
Plant Protection ◽  
Point Clouds ◽  
Biomass Estimation ◽  
Reconstruction Method ◽  
Volume Calculation ◽  
Alpha Shape ◽  
3D Point Clouds ◽  
Canopy Volume

The characteristic parameters of Citrus grandis var. Longanyou canopies are important when measuring yield and spraying pesticides. However, the feasibility of the canopy reconstruction method based on point clouds has not been confirmed with these canopies. Therefore, LiDAR point cloud data for C. grandis var. Longanyou were obtained to facilitate the management of groves of this species. Then, a cloth simulation filter and European clustering algorithm were used to realize individual canopy extraction. After calculating canopy height and width, canopy reconstruction and volume calculation were realized using six approaches: by a manual method and using five algorithms based on point clouds (convex hull, CH; convex hull by slices; voxel-based, VB; alpha-shape, AS; alpha-shape by slices, ASBS). ASBS is an innovative algorithm that combines AS with slices optimization, and can best approximate the actual canopy shape. Moreover, the CH algorithm had the shortest run time, and the R2 values of VCH, VVB, VAS, and VASBS algorithms were above 0.87. The volume with the highest accuracy was obtained from the ASBS algorithm, and the CH algorithm had the shortest computation time. In addition, a theoretical but preliminarily system suitable for the calculation of the canopy volume of C. grandis var. Longanyou was developed, which provides a theoretical reference for the efficient and accurate realization of future functional modules such as accurate plant protection, orchard obstacle avoidance, and biomass estimation.

Multi-Radius Density Clustering Algorithm Based on Outlier Factor

2014 ◽  
Vol 472 ◽  
pp. 427-431
Author(s):  
Zong Lin Ye ◽  
Hui Cao ◽  
Li Xin Jia ◽  
Yan Bin Zhang ◽  
Gang Quan Si
Keyword(s):  
Clustering Algorithm ◽  
Spatial Clustering ◽  
Similar Process ◽  
The Core ◽  
Dbscan Algorithm ◽  
Proposed Model ◽  
Density Clustering ◽  
Relationship Of ◽  
Core Points ◽  
The Relationship

This paper proposes a novel multi-radius density clustering algorithm based on outlier factor. The algorithm first calculates the density-similar-neighbor-based outlier factor (DSNOF) for each point in the dataset according to the relationship of the density of the point and its neighbors, and then treats the point whose DSNOF is smaller than 1 as a core point. Second, the core points are used for clustering by the similar process of the density based spatial clustering application with noise (DBSCAN) to get some sub-clusters. Third, the proposed algorithm merges the obtained sub-clusters into some clusters. Finally, the points whose DSNOF are larger than 1 are assigned into these clusters. Experiments are performed on some real datasets of the UCI Machine Learning Repository and the experiments results verify that the effectiveness of the proposed model is higher than the DBSCAN algorithm and k-means algorithm and would not be affected by the parameter greatly.

A novel image segmentation method based on fast density clustering algorithm

2018 ◽  
Vol 73 ◽  
pp. 92-110 ◽  
Author(s):  
Jinyin Chen ◽  
Haibin Zheng ◽  
Xiang Lin ◽  
Yangyang Wu ◽  
Mengmeng Su
Keyword(s):  
Image Segmentation ◽  
Clustering Algorithm ◽  
Segmentation Method ◽  
Density Clustering

scholarly journals Processing Chain for Estimation of Tree Diameter from GNSS-IMU-Based Mobile Laser Scanning Data

2019 ◽  
Vol 11 (6) ◽  
pp. 615 ◽  
Author(s):  
Juraj Čerňava ◽  
Martin Mokroš ◽  
Ján Tuček ◽  
Michal Antal ◽  
Zuzana Slatkovská
Keyword(s):  
Point Cloud ◽  
Laser Scanning ◽  
Forest Canopy ◽  
Point Clouds ◽  
Satellite System ◽  
Forest Road ◽  
Forest Mapping ◽  
Fitting Method ◽  
Progressive Technology ◽  
Global Navigation Satellite

Mobile laser scanning (MLS) is a progressive technology that has already demonstrated its ability to provide highly accurate measurements of road networks. Mobile innovation of the laser scanning has also found its use in forest mapping over the last decade. In most cases, existing methods for forest data acquisition using MLS result in misaligned scenes of the forest, scanned from different views appearing in one point cloud. These difficulties are caused mainly by forest canopy blocking the global navigation satellite system (GNSS) signal and limited access to the forest. In this study, we propose an approach to the processing of MLS data of forest scanned from different views with two mobile laser scanners under heavy canopy. Data from two scanners, as part of the mobile mapping system (MMS) Riegl VMX-250, were acquired by scanning from five parallel skid trails that are connected to the forest road. Misaligned scenes of the forest acquired from different views were successfully extracted from the raw MLS point cloud using GNSS time based clustering. At first, point clouds with correctly aligned sets of ground points were generated using this method. The loss of points after the clustering amounted to 33.48%. Extracted point clouds were then reduced to 1.15 m thick horizontal slices, and tree stems were detected. Point clusters from individual stems were grouped based on the diameter and mean GNSS time of the cluster acquisition. Horizontal overlap was calculated for the clusters from individual stems, and sufficiently overlapping clusters were aligned using the OPALS ICP module. An average misalignment of 7.2 mm was observed for the aligned point clusters. A 5-cm thick horizontal slice of the aligned point cloud was used for estimation of the stem diameter at breast height (DBH). DBH was estimated using a simple circle-fitting method with a root-mean-square error of 3.06 cm. The methods presented in this study have the potential to process MLS data acquired under heavy forest canopy with any commercial MMS.

Sign in / Sign up

Export Citation Format

Share Document