Article
Environmental Sciences
Philipp Glira, Christoph Weidinger, Johannes Otepka-Schremmer, Camillo Ressl, Norbert Pfeifer, Michaela Haberler-Weber
Summary: Nonrigid registration is a significant challenge in point cloud processing with diverse applications. This paper presents a new method using piecewise tricubic polynomials to model nonrigid deformations, offering several advantages over existing methods.
Article
Geochemistry & Geophysics
Leping He, Shuaiqing Wang, Qijun Hu, Qijie Cai, Muyao Li, Yu Bai, Kai Wu, Bo Xiang
Summary: This article proposes a new method for three-dimensional point cloud registration, optimizing the geometric features to improve accuracy and speed. Experimental results demonstrate that this method significantly outperforms traditional ICP and other state-of-the-art registration methods in terms of accuracy and speed.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Optics
Song Zhang, Qifeng Wang, Jianxin Zhang, Bin Liu
Summary: This study presents a learning-based registration method for sub models of mesh models obtained by optical scanning. It addresses the limitations of current registration methods in achieving sub model registration under poor initial position conditions and common adverse conditions.
OPTICS AND LASER TECHNOLOGY
(2023)
Article
Chemistry, Multidisciplinary
Haiyuan Cao, Deng Chen, Zhaohui Zheng, Yanduo Zhang, Huabing Zhou, Jianping Ju
Summary: Point cloud registration plays a crucial role in various applications such as 3D reconstruction and pose estimation. However, traditional ICP algorithm has limitations in terms of dependence on initial position, robustness, and timeliness. To address these issues, a fast point cloud registration method that incorporates RGB image information is proposed. This method utilizes SIFT algorithm for feature point detection, RANSAC algorithm for outlier removal and initial transformation matrix calculation, and FR-ICP algorithm for precise registration.
APPLIED SCIENCES-BASEL
(2023)
Article
Geography, Physical
Jiayuan Li, Qingwu Hu, Yongjun Zhang, Mingyao Ai
Summary: This paper proposes a robust symmetric ICP (RSICP) algorithm to address the limitations of traditional ICP, including small convergence basin and sensitivity to outliers and partial overlaps. The algorithm introduces a new symmetric point-to-plane distance metric and an adaptive robust loss, as well as a simple and effective linearization method. Extensive experiments demonstrate that the proposed algorithm outperforms other methods in terms of both accuracy and efficiency.
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
(2022)
Article
Chemistry, Multidisciplinary
Ruiyang Sun, Enzhong Zhang, Deqiang Mu, Shijun Ji, Ziqiang Zhang, Hongwei Liu, Zheng Fu
Summary: This paper proposes a new point cloud registration method that optimizes the rough registration and precise registration stages. By improving the feature point extraction and point cloud filtering methods, and introducing the voxel concept for point cloud filtering, experimental results show noise removal rates of 95.3%, 98.6%, and 93.5%. In the precise registration stage, a method combining curvature feature and fast point feature histogram is proposed and analyzed experimentally. The analysis and verification of datasets such as Stanford bunny and free-form surface show approximately 40.16% and 36.27% reduction in error, and approximately 42.9% and 37.14% improvement in iteration times.
APPLIED SCIENCES-BASEL
(2023)
Article
Chemistry, Analytical
Marco A. Barreto, Jorge Perez-Gonzalez, Hugh M. Herr, Joel C. Huegel
Summary: This work aims to design, prototype, and test a functional system for scanning plaster cast molds, which could be used for lower limb reconstruction applications. The image capture system, consisting of 5 cameras and algorithms, shows good segmentation metrics and visual similarity with the actual images.
Article
Computer Science, Hardware & Architecture
Rui Guo, Jinqian Chen, Lin Wang
Summary: Multi-view registration is a significant research issue in computer vision and robotics, with most existing works focusing on point to point match correspondence but facing limitations. To address these limitations, a novel Hierarchical K-means Clustering Registration method is proposed, which achieves more robust and accurate results by gradually increasing the number of clusters. Extensive experiments demonstrate the effectiveness and robustness of the proposed method compared to state-of-the-art methods.
COMPUTERS & ELECTRICAL ENGINEERING
(2021)
Article
Environmental Sciences
Xiangxiong Kong
Summary: This study introduces a novel cliff monitoring methodology that does not rely on georeferencing efforts, producing reliable monitoring results by processing 3D point clouds and aligning them using a rigid registration protocol. Experimental findings demonstrate the efficiency of this approach in small-scale experiments and full-scale field validation, highlighting its significance for underserved coastal communities.
Article
Mathematics
Artyom Makovetskii, Sergei Voronin, Vitaly Kober, Alexei Voronin
Summary: Point cloud collection for 3D scene formation often requires using information from multiple data scans. The common approach is to register point cloud pairs consecutively using the iterative closest point (ICP) algorithm, but most versions of the ICP algorithm are only suitable for small movements between two point clouds, making it difficult to accumulate multiple scans. A modified algorithm that combines the ICP and RANSAC concepts is proposed in this paper, which automatically selects model parameters and can be parallelized. The performance of the proposed algorithm is compared with known global registration algorithms.
Article
Engineering, Marine
Y. Argouarc'h, R. Creac'hcadec
Summary: The three-dimensional geometrical evolution of mooring chain links exposed to the marine environment can be observed using 3D scanners. Precise identification of material loss caused by corrosion or fretting is possible through successive surveys. However, alignment of obtained scans is necessary due to differences in reference marks. An alignment algorithm based on the iterative closest point method (ICP) is developed to address the issue. Tests conducted on CAD constructions show that the algorithm achieves sufficient precision for the objectives of the work.
Article
Chemistry, Analytical
Li Zheng, Zhukun Li
Summary: This research proposes a method for multi-source point cloud data fusion based on Fast Point Feature Histograms (FPFH) feature difference, which can improve the accuracy of low-precision point cloud by generating virtual point pairs and achieving more accurate registration in the ICP algorithm.
Article
Optics
Meiting Xin, Bing Li, Xiang Wei, Zhuo Zhao
Summary: This paper proposes a rapid 3D point cloud registration method based on weighted principal component analysis and re-weighted iterative closest point algorithm, which consists of three stages: coarse alignment, data simplification, and fine alignment, aiming to improve the efficiency of reverse engineering.
Article
Computer Science, Information Systems
Anjana Puri, Abeer Alsadoon, P. W. C. Prasad, Israa Al-Neami, Sami Haddad
Summary: In this study, a Correntropy based scale ICP algorithm is proposed to improve image registration during jaw surgery. The proposed system includes Enhanced Tracking Learning Detection (TLD) and a Modified Correntropy-based enhanced ICP (MCbeICP) algorithm, which enhance registration accuracy and reduce processing time.
MULTIMEDIA TOOLS AND APPLICATIONS
(2022)
Article
Remote Sensing
Yongzhi Wang, Tao Zhou, Hui Li, Wenlong Tu, Jing Xi, Lixia Liao
Summary: This article proposes an improved ICP registration algorithm for 3D lidar point cloud data. By using the GMM method and corner features, the registration accuracy and efficiency are improved. Experimental results show that this method outperforms other traditional registration methods in terms of accuracy and efficiency, and effectively solves the problem of local optima.
INTERNATIONAL JOURNAL OF REMOTE SENSING
(2022)