A new method for building roof segmentation from airborne LiDAR point cloud data

Kong, Deming; Xu, Lijun; Li, Xiaolu

doi:10.1088/0957-0233/24/9/095402

Cited by 20 publications

(9 citation statements)

References 27 publications

(36 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Kong et al proposed a segmentation method based on k-plane clustering [20], which has less complexity and higher accuracy than fuzzy k-means [21]. Kong et al combined the k-plane and k-means algorithms for roof segmentation, where the k-plane algorithm was introduced to make up for the deficiency of the k-means algorithm, so that high-precision segmentation could be achieved [22]. To better calculate the features for every point, Filin and Pfeifer utilized a slope-adaptive neighborhood system to compute the attributes of the points, and then applied a mode-seeking algorithm to extract clusters, considering the spatial positions and the normal vectors of the points [23].…”

Section: Feature Clustering Based Methodsmentioning

confidence: 99%

Roof Plane Segmentation From LiDAR Point Cloud Data Using Region Expansion Based L ₀ Gradient Minimization and Graph Cut

Wang¹,

Ji²

2021

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Automatic roof segmentation from airborne light detection and ranging (LiDAR) point cloud data is a key technology for building reconstruction and digital city modeling. In this paper, we develop a novel region expansion based L0 gradient minimization algorithm for processing unordered point cloud data, and a two-stage global optimization method consisting of the L0 gradient minimization and graph cut for roof plane segmentation. Firstly, we extract the LiDAR points of buildings from the original point cloud data with a deep learning based method and separate the points of the different buildings using Euclidean clustering to improve the processing efficiency. Secondly, region expansion based L0 gradient minimization is proposed, which is specially designed for roof plane segmentation from unordered point clouds. To fundamentally avoid the need for empirical parameter tuning in L0 gradient minimization, we propose a multi-stage coarseto-fine segmentation process, which further improves the effect of the roof plane segmentation. Finally, graph cut is utilized to solve the jagged boundary and over-segmentation problems existing in the segmented roof planes and produce the segmentation results. We conducted comparative experiments on the Vaihingen and Hangzhou datasets. The experimental results show that the proposed approach significantly outperforms the current state-of-the-art approaches at least 6.7% and 8.9% in roof plane quality index in the Vaihingen and Hangzhou datasets, while showing superior robustness to different kinds of data.

show abstract

Section: Feature Clustering Based Methodsmentioning

confidence: 99%

Roof Plane Segmentation From LiDAR Point Cloud Data Using Region Expansion Based L ₀ Gradient Minimization and Graph Cut

Wang¹,

Ji²

2021

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

show abstract

“…The representative methods are region growing plane segmentation algorithm [22], which uses the smoothness factor to extract surfaces, M-estimator Sample and Consensus [23], which repeatedly fits planes with randomly selected points until generating a proper surface, element shape detection [24], which segments points based on element shape (i.e., walls are vertical surfaces), the fusion of color and spatial data [25], which combines the color features and spatial features together, etc. For the machine learning-based method, there are two popular algorithms: k-means clustering [26,27], which initially defines a set of center points and then updates the centers based on point distributions, and Hierarchical clustering [28,29], which initializes each point as a group and then hierarchically increases the groups according to the distances between points.…”

Section: A State-of-the-art Research On Bim Developmentmentioning

confidence: 99%

An Integrated Framework for BIM Development of Concrete Buildings Containing Both Surface Elements and Rebar

Xiang

Rashidi

2023

IEEE Access

View full text Add to dashboard Cite

As-built Building Information Models (BIM) are increasingly used to facilitate the management of all aspects of built infrastructure's life cycle. Existing studies mainly focus on automating as-built BIM development for surface elements but often ignore embedded elements such as rebar due to the inaccessibility with typical sensing devices, such as image-based or time-of-flight-based methods. To tackle the issue, this research innovatively utilizes Ground Penetrating Radar (GPR) and the photogrammetry method together to generate BIMs for in-service buildings considering both surface elements (e.g., column, slab, wall) and rebar. As-built BIM for surface elements is developed at first and rebar is identified by using GPR. A calibration label is designed and attached to elements which are scanned by GPR device, and a series of images are captured from those elements and then used with other images to generate point clouds. Faster RCNN is utilized to recognize labels among all images. Then, an inverse photogrammetry approach is deployed to identify the scanned elements in BIM. By matching the recorded timestamps of GPR data and labeled images, links between the rebar in GPR data and elements in BIMs are successfully developed. Finally, 𝐼𝐹𝐶 is developed to generate as-built BIM models. Six cases studies demonstrate the system provides a highly automated procedure to develop as-built BIM, and rebars could be efficiently localized and projected into corresponding elements in BIM.INDEX TERMS As-built BIM, deep learning, GPR, IFC, inverse photogrammetry, point cloud, rebar placement.

show abstract

“…3D point cloud data, as a dense collection of points used to depict the surface characteristics of the target, is structureless. Consequently, the extraction of planar structures from 3D laser point clouds has been made a new direction of research on computer vision [ 6 , 7 , 8 , 9 ].…”

Section: Introductionmentioning

confidence: 99%

High-Precision Plane Detection Method for Rock-Mass Point Clouds Based on Supervoxel

Yu¹,

Xiao²,

Wang³

2020

Sensors

View full text Add to dashboard Cite

In respect of rock-mass engineering, the detection of planar structures from the rock-mass point clouds plays a crucial role in the construction of a lightweight numerical model, while the establishment of high-quality models relies on the accurate results of surface analysis. However, the existing techniques are barely capable to segment the rock mass thoroughly, which is attributed to the cluttered and unpredictable surface structures of the rock mass. This paper proposes a high-precision plane detection approach for 3D rock-mass point clouds, which is effective in dealing with the complex surface structures, thus achieving a high level of detail in detection. Firstly, the input point cloud is fast segmented to voxels using spatial grids, while the local coplanarity test and the edge information calculation are performed to extract the major segments of planes. Secondly, to preserve as much detail as possible, supervoxel segmentation instead of traditional region growing is conducted to deal with scattered points. Finally, a patch-based region growing strategy applicable to rock mass is developed, while the completed planes are obtained by merging supervoxel patches. In this paper, an artificial icosahedron point cloud and four rock-mass point clouds are applied to validate the performance of the proposed method. As indicated by the experimental results, the proposed method can make high-precision plane detection achievable for rock-mass point clouds while ensuring high recall rate. Furthermore, the results of both qualitative and quantitative analyses evidence the superior performance of our algorithm.

show abstract

A new method for building roof segmentation from airborne LiDAR point cloud data

Cited by 20 publications

References 27 publications

Roof Plane Segmentation From LiDAR Point Cloud Data Using Region Expansion Based L ₀ Gradient Minimization and Graph Cut

Roof Plane Segmentation From LiDAR Point Cloud Data Using Region Expansion Based L ₀ Gradient Minimization and Graph Cut

An Integrated Framework for BIM Development of Concrete Buildings Containing Both Surface Elements and Rebar

High-Precision Plane Detection Method for Rock-Mass Point Clouds Based on Supervoxel

Contact Info

Product

Resources

About

A new method for building roof segmentation from airborne LiDAR point cloud data

Cited by 20 publications

References 27 publications

Roof Plane Segmentation From LiDAR Point Cloud Data Using Region Expansion Based L 0 Gradient Minimization and Graph Cut

Roof Plane Segmentation From LiDAR Point Cloud Data Using Region Expansion Based L 0 Gradient Minimization and Graph Cut

An Integrated Framework for BIM Development of Concrete Buildings Containing Both Surface Elements and Rebar

High-Precision Plane Detection Method for Rock-Mass Point Clouds Based on Supervoxel

Contact Info

Product

Resources

About

Roof Plane Segmentation From LiDAR Point Cloud Data Using Region Expansion Based L ₀ Gradient Minimization and Graph Cut

Roof Plane Segmentation From LiDAR Point Cloud Data Using Region Expansion Based L ₀ Gradient Minimization and Graph Cut