Evaluation of the ICP Algorithm in 3D Point Cloud Registration

Li, Peng; Wang, Rui-Sheng; Wang, Yanxia; Tao, Wuyong

doi:10.1109/access.2020.2986470

Cited by 121 publications

(54 citation statements)

References 94 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The second adjusts the point cloud through the minimization of the distance between the point in the point cloud of reference and the tangent plane to its corresponding point in the point cloud to register. Studies shown that the validity of point-to-point ICP is more robust to Gaussian noise than point-to-plane ICP [60]. However, the latter undergoes fewer iterations, and it is more effective [60].…”

Section: Alignmentmentioning

confidence: 99%

“…Studies shown that the validity of point-to-point ICP is more robust to Gaussian noise than point-to-plane ICP [60]. However, the latter undergoes fewer iterations, and it is more effective [60]. Since the point clouds are cleaned and filtered from noise before starting the registration process, the point-to-point ICP method was chosen for this work.…”

Section: Alignmentmentioning

confidence: 99%

See 1 more Smart Citation

Automatic Point-Cloud Registration for Quality Control in Building Works

et al. 2021

View full text Add to dashboard Cite

Total and automatic digitalization of indoor spaces in 3D implies a great advance in building maintenance and construction tasks, which currently require visits and manual works. Terrestrial laser scanners (TLS) have been widely used for these tasks, although the acquisition methodology with TLS systems is time consuming, and each point cloud is acquired in a different coordinate system, so the user has to post-process the data to clean and get a unique point cloud of the whole scenario. This paper presents a solution for the automatic data acquisition and registration of point clouds from indoor scenes, designed for point clouds acquired with a terrestrial laser scanner (TLS) mounted on an unmanned ground vehicle (UGV). The methodology developed allows the generation of one complete dense 3D point cloud consisting of the acquired point clouds registered in the same coordinate system, reaching an accuracy below 1 cm in section dimensions and below 1.5 cm in walls thickness, which makes it valid for quality control in building works. Two different study cases corresponding to building works were chosen for the validation of the method, showing the applicability of the methodology developed for tasks related to the control of the evolution of the construction.

show abstract

Section: Alignmentmentioning

confidence: 99%

Section: Alignmentmentioning

confidence: 99%

Automatic Point-Cloud Registration for Quality Control in Building Works

et al. 2021

View full text Add to dashboard Cite

show abstract

“…When a suitable model is present, model-driven building extraction is more robust (Li et al, 2020). For example, Karantzalos and Paragios (2008) proposed a set of building templates as shape priors to guide the segmentation of buildings from aerial imagery, achieving superior results to pure intensitybased segmentation.…”

Section: Building Extraction For Construction Sitesmentioning

confidence: 99%

Multi-Building Extraction and Alignment for As-Built Point Clouds: A Case Study With Crane Cameras

Masood

Aikala

Seppänen

et al. 2020

Front. Built Environ.

View full text Add to dashboard Cite

Automatic reality capture and monitoring of construction sites can reduce costs, accelerate timelines and improve quality in construction projects. Recently, automatic close-range capture of the state of large construction sites has become possible through crane and drone-mounted cameras, which results in sizeable, noisy, multi-building as-built point clouds. To infer construction progress from these point clouds, they must be aligned with the as-designed BIM model. Unlike the problem of aligning single buildings, the multi-building scenario is not well-studied. In this work, we address some unique issues that arise in the alignment of multi-building point clouds. Firstly, we show that a BIM-based 3D filter is a versatile tool that can be used at multiple stages of the alignment process. We use the building-pass filter to remove non-building noise and thus extract the buildings, delineate the boundaries of the building after the base is identified and as a post-processing step after the alignment is achieved. Secondly, in light of the sparseness of some buildings due to partial capture, we propose to use the best-captured building as a pivot to align the entire point cloud. We propose a fully automated three-step alignment process that leverages the simple geometry of the pivot building and aligns partial xy-projections, identifies the base using z-histograms and aligns the bounding boxes of partial yz-projections. Experimental results with crane camera point clouds of a large construction site show that our proposed techniques are fast and accurate, allowing us to estimate the current floor under construction from the aligned clouds and enabling potential slab state analysis. This work contributes a fully automated method of reality capture and monitoring of multi-building construction sites.

show abstract

“…Point cloud registration, e.g., the Iterative Closest Point (ICP) algorithm, aligns the point clouds to compute the rigid pose transformation. Despite its simplicity and efficiency, ICP requires a good initialization and is easily trapped in the local optimum [12]. Previous efforts have been made to improve ICP for the global optimal solution [13][14][15].…”

Section: Introductionmentioning

confidence: 99%

A Coarse-to-Fine Method for Estimating the Axis Pose Based on 3D Point Clouds in Robotic Cylindrical Shaft-in-Hole Assembly

Can

Chen

et al. 2021

Sensors

View full text Add to dashboard Cite

In this work, we propose a novel coarse-to-fine method for object pose estimation coupled with admittance control to promote robotic shaft-in-hole assembly. Considering that traditional approaches to locate the hole by force sensing are time-consuming, we employ 3D vision to estimate the axis pose of the hole. Thus, robots can locate the target hole in both position and orientation and enable the shaft to move into the hole along the axis orientation. In our method, first, the raw point cloud of a hole is processed to acquire the keypoints. Then, a coarse axis is extracted according to the geometric constraints between the surface normals and axis. Lastly, axis refinement is performed on the coarse axis to achieve higher precision. Practical experiments verified the effectiveness of the axis pose estimation. The assembly strategy composed of axis pose estimation and admittance control was effectively applied to the robotic shaft-in-hole assembly.

show abstract

Evaluation of the ICP Algorithm in 3D Point Cloud Registration

Cited by 121 publications

References 94 publications

Automatic Point-Cloud Registration for Quality Control in Building Works

Automatic Point-Cloud Registration for Quality Control in Building Works

Multi-Building Extraction and Alignment for As-Built Point Clouds: A Case Study With Crane Cameras

A Coarse-to-Fine Method for Estimating the Axis Pose Based on 3D Point Clouds in Robotic Cylindrical Shaft-in-Hole Assembly

Contact Info

Product

Resources

About