Iterative Closest Point Algorithm Research Articles

Wheeled Robot Visual Odometer Based on Two-dimensional Iterative Closest Point Algorithm

According to the two-dimensional motion characteristics of planar motion wheeled robot, the visual odometer was dimensionally reduced in this study. In the feature point matching part of visual odometer, the contour constraint was used to filter out the mismatched feature point pairs (abbreviated as FPP). This method could also filter out the matched FPP, and the feature of FPP was correct color image matches, however, their depth image error was large. This offered higher quality matched FPP for the subsequent interframe motion estimation. Dimension reduction was performed in the interframe motion estimation part, and the two-dimensional Iterative Closest Point (ICP) algorithm was used for camera motion estimation. The experiments indicated that the proposed algorithm effectively improved the computational speed and precision of planar motion wheeled robot visual odometer. This research indicates that the dimension reduction processing of ICP algorithm can effectively improve the operation speed and calculation accuracy of planar motion wheeled robot visual odometry, which provides a good reference and data support for the subsequent research of wheeled robot visual odometry in the future.

Research on Side-Slope Monitoring by Integrating Terrestrial Laser Scanning and UAV-Based Photogrammetry

Abstract Side-slope deformation monitoring compares monitoring data from the same area over different periods and measures the deformation variables. Because of the gaps and coarseness of side-slope monitoring data, a side-slope monitoring method that integrates terrestrial laser scanning (TLS) and unmanned aerial vehicle (UAV)–based photogrammetry point clouds is proposed, aiming to solve the problem of slope monitoring in complex scenes. First, TLS and UAV-based photogrammetry point clouds are acquired. Then, the two types of point clouds are registered by an iterative closest point algorithm. Next, the data gap areas in the TLS point cloud are detected, and a gap-filling method is used to integrate the UAV-based photogrammetry point cloud with the TLS point cloud. Finally, side-slope deformation is detected based on a multiscale model-to-model cloud comparison algorithm. A side slope in Chenggong, Kunming, China, is taken as an example. The surface deformation of the side slope was monitored during January and June 2021. The experimental results show that the registration errors of the two-phase integration point cloud are 0.039 m and 0.035 m. The root mean square errors of the four ground checkpoints are 0.033 m and 0.038 m. Finally, the side slope is found to have deformed and formed a main deformation area, which shows that this side slope was in an active state.

ICP Algorithm: Theory, Practice and Its SLAM-oriented Taxonomy

The Iterative Closest Point (ICP) algorithm is one of the most important algorithms for ge-ometric alignment of three-dimensional surface registration, which is frequently used in computer vision tasks, including the Simultaneous Localization And Mapping (SLAM) tasks. In this paper, we illustrate the theoretical principles of the ICP algorithm, how it can be used in surface registration tasks, and the traditional taxonomy of the variants of the ICP algorithm. As SLAM is becoming a popular topic, we also introduce a SLAM-oriented taxonomy of the ICP algorithm, based on the characteristics of each type of SLAM task, including whether the SLAM task is online or not and whether the landmarks are present as features in the SLAM task. We make a synthesis of each type of SLAM task by compar-ing several up-to-date research papers and analyzing their implementation details.

ICP Algorithm: Theory, Practice and Its SLAM-oriented Taxonomy

The Iterative Closest Point (ICP) algorithm is one of the most important algorithms for ge-ometric alignment of three-dimensional surface registration, which is frequently used in computer vision tasks, including the Simultaneous Localization And Mapping (SLAM) tasks. In this paper, we illustrate the theoretical principles of the ICP algorithm, how it can be used in surface registration tasks, and the traditional taxonomy of the variants of the ICP algorithm. As SLAM is becoming a popular topic, we also introduce a SLAM-oriented taxonomy of the ICP algorithm, based on the characteristics of each type of SLAM task, including whether the SLAM task is online or not and whether the landmarks are present as features in the SLAM task. We make a synthesis of each type of SLAM task by compar-ing several up-to-date research papers and analyzing their implementation details.

Optimization of the 3D Point Cloud Registration Algorithm Based on FPFH Features

In order to solve the problem of the traditional iterative closest point algorithm (ICPA), which requires a high initial position of point cloud and improves the speed and accuracy of point cloud registration, a new registration method is proposed in this paper. Firstly, the rough registration method is optimized. As for the extraction of the feature points, a new method of feature point extraction is adopted, which can better keep the features of the original point cloud. At the same time, the traditional point cloud filtering method is improved, and a voxel idea is introduced to filter the point cloud. The edge length data of the voxels is determined by the density, and the experimentally verified noise removal rates for the 3D cloud data are 95.3%, 98.6%, and 93.5%, respectively. Secondly, a precise registration method that combines the curvature feature and fast point feature histogram (FPFH) is proposed in the precise registration stage, and the algorithm is analyzed experimentally. Finally, the two point cloud data sets Stanford bunny and free-form surface are analyzed and verified, and it is concluded that this method can reduce the error by about 40.16% and 36.27%, respectively, and improve the iteration times by about 42.9% and 37.14%, respectively.

A Chebyshev metamodel based BnB approach to efficiently search global optimum for 3D ICP point set registration

Iterative Closest Point (ICP) algorithm for 3D point set registration has many robust properties that make it widely used. Regardless of the initialization, the Go-ICP method was first presented to search global optimum for 3D point set registration using Branch-and-Bound (BnB) scheme, but it could be time consuming. In this paper, an efficient modification, called Ego-ICP, is presented. The core idea is: (1) SE(3) feasible domain, L2-norm registration model of ICP and corresponding local ICP algorithm are reviewed, (2) Chebyshev metamodel is implemented to approximate the registration model with specific rotation, (3) the Chebyshev metamodel based single nested BnB approach is then presented to efficiently search global optimum for 3D point set registration. Furthermore, three experimental targets for different Datasets are demonstrated and shown that the presented approach, as compared with the BNB based method, the geometry based methods and the deep learning based method, (1) can be faster than or close to the efficiency of other compared methods, (2) can be reliable regardless of the initialization, and (3) can be 100% successful even if noise and outlier are existed. Lastly, several insights are provided to further improve the efficiency of our method.

A Multi-Resolution Approach to Point Cloud Registration without Control Points

Terrestrial photographic imagery combined with structure-from-motion (SfM) provides a relatively easy-to-implement method for monitoring environmental systems, even in remote and rough terrain. However, the collection of in-situ positioning data and the identification of control points required for georeferencing in SfM processing is the primary roadblock to using SfM in difficult-to-access locations; it is also the primary bottleneck for using SfM in a time series. We describe a novel, computationally efficient, and semi-automated approach for georeferencing unreferenced point clouds (UPC) derived from terrestrial overlapping photos to a reference dataset (e.g., DEM or aerial point cloud; hereafter RPC) in order to address this problem. The approach utilizes a Discrete Global Grid System (DGGS), which allows us to capitalize on easily collected rough information about camera deployment to coarsely register the UPC using the RPC. The DGGS also provides a hierarchical set of grids which supports a hierarchical modified iterative closest point algorithm with natural correspondence between the UPC and RPC. The approach requires minimal interaction in a user-friendly interface, while allowing for user adjustment of parameters and inspection of results. We illustrate the approach with two case studies: a close-range (<1 km) vertical glacier calving front reconstructed from two cameras at Fountain Glacier, Nunavut and a long-range (>3 km) scene of relatively flat glacier ice reconstructed from four cameras overlooking Nàłùdäy (Lowell Glacier), Yukon, Canada. We assessed the accuracy of the georeferencing by comparing the UPC to the RPC, as well as surveyed control points; the consistency of the registration was assessed using the difference between successive registered surfaces in the time series. The accuracy of the registration is roughly equal to the ground sampling distance and is consistent across time steps. These results demonstrate the promise of the approach for easy-to-implement georeferencing of point clouds from terrestrial imagery with acceptable accuracy, opening the door for new possibilities in remote monitoring for change-detection, such as monitoring calving rates, glacier surges, or other seasonal changes at remote field locations.

Point Cloud Registration Methods for Long-Span Bridge Spatial Deformation Monitoring Using Terrestrial Laser Scanning

In recent years, efforts have been devoted to utilizing terrestrial laser scanning for bridge spatial performance inspection, but they are still restricted to small or medium-span bridges, like some historical heritages. Due to the large-scale dimensional features of long-span bridges, applications of 3D point cloud techniques still remain challenging, such as the extra-long scan range and extreme-small incidence angle when scanning a bridge with a span over 1000 m. Moreover, rare attempts can be found for the performance evaluation of point cloud registration methods for long-span bridges as well, which is a critical basis for further spatial deformation recognition on the point cloud data. Hence, in this study, a cross-evaluation of three iterative closest point (ICP) registration methods is conducted for long-span suspension bridges, namely, traditional ICP, kd-tree-based ICP, and feature point-based ICP algorithms. We conducted field laser scanning on the Ma’anshan Yangtze Bridge, an 1880 m long suspension bridge located in China. The results show that the feature point-based ICP algorithm outperforms the other two in terms of convergence rate and execution time for a single iteration due to the smaller number of registration points compared to the other two algorithms. Moreover, it also gives more precise values in terms of bridge tower spatial deformation identification. Meanwhile, due to efficient data organization, the kd-tree-based ICP algorithm takes less time for a single iteration than the traditional ICP algorithm. Finally, two suggestions for algorithm improvement in terms of efficiency optimization and accuracy improvement of long-span bridge deformation analysis are proposed based on the assessment results.

Accelerated point set registration method

In computer vision and robotics, point set registration is a fundamental issue used to estimate the relative position and orientation (pose) of an object in an environment. In a rapidly changing scene, this method must be executed frequently and in a timely manner, or the pose estimation becomes outdated. The point registration method is a computational bottleneck of a vision-processing pipeline. For this reason, this paper focuses on speeding up a widely used point registration method, the iterative closest point (ICP) algorithm. In addition, the ICP algorithm is transformed into a massively parallel algorithm and mapped onto a vector processor to realize a speedup of approximately an order of magnitude. Finally, we provide algorithmic and run-time analysis.

Building Maps Using Monocular Image-feeds from Windshield-mounted Cameras in a Simulator Environment

3-dimensional, accurate, and up-to-date maps are essential for vehicles with autonomous capabilities, whose functionality is made possible by machine learning-based algorithms. Since these solutions require a tremendous amount of data for parameter optimization, simulation-to-reality (Sim2Real) methods have been proven immensely useful for training data generation. For creating realistic models to be used for synthetic data generation, crowdsourcing techniques present a resource-efficient alternative. In this paper, we show that using the Carla simulation environment, a crowdsourcing model can be created that mimics a multi-agent data gathering and processing pipeline. We developed a solution that yields dense point clouds based on monocular images and location information gathered by individual data acquisition vehicles. Our method provides scene reconstructions using the robust Structure-from-Motion (SfM) solution of Colmap. Moreover, we introduce a solution for synthesizing dense ground truth point clouds originating from the Carla simulator using a simulated data acquisition pipeline. We compare the results of the Colmap reconstruction with the reference point cloud after aligning them using the iterative closest point algorithm. Our results show that a precise point cloud reconstruction was feasible with this crowdsourcing-based approach, with 54\% of the reconstructed points having an error under 0.05 m, and a weighted root mean square error of 0.0449 m for the entire point cloud.

Dynamic detection of three-dimensional crop phenotypes based on a consumer-grade RGB-D camera.

Nondestructive detection of crop phenotypic traits in the field is very important for crop breeding. Ground-based mobile platforms equipped with sensors can efficiently and accurately obtain crop phenotypic traits. In this study, we propose a dynamic 3D data acquisition method in the field suitable for various crops by using a consumer-grade RGB-D camera installed on a ground-based movable platform, which can collect RGB images as well as depth images of crop canopy sequences dynamically. A scale-invariant feature transform (SIFT) operator was used to detect adjacent date frames acquired by the RGB-D camera to calculate the point cloud alignment coarse matching matrix and the displacement distance of adjacent images. The data frames used for point cloud matching were selected according to the calculated displacement distance. Then, the colored ICP (iterative closest point) algorithm was used to determine the fine matching matrix and generate point clouds of the crop row. The clustering method was applied to segment the point cloud of each plant from the crop row point cloud, and 3D phenotypic traits, including plant height, leaf area and projected area of individual plants, were measured. We compared the effects of LIDAR and image-based 3D reconstruction methods, and experiments were carried out on corn, tobacco, cottons and Bletilla striata in the seedling stage. The results show that the measurements of the plant height (R²= 0.9~0.96, RSME = 0.015~0.023 m), leaf area (R²= 0.8~0.86, RSME = 0.0011~0.0041 m 2 ) and projected area (R² = 0.96~0.99) have strong correlations with the manual measurement results. Additionally, 3D reconstruction results with different moving speeds and times throughout the day and in different scenes were also verified. The results show that the method can be applied to dynamic detection with a moving speed up to 0.6m/s and can achieve acceptable detection results in the daytime, as well as at night. Thus, the proposed method can improve the efficiency of individual crop 3D point cloud data extraction with acceptable accuracy, which is a feasible solution for crop seedling 3D phenotyping outdoors.

Three-dimensional point cloud registration algorithm based on the correlation coefficient

Point cloud registration is an important research problem in machine vision. One of the most widely used methods for point set registration is the iterative closest point (ICP) algorithm. However, ICP is known to be susceptible to local minima because it is based on the local iterative optimization technique and its performance critically relies on the quality of the initialization. To reduce the possibility of the algorithm falling into local minima, we propose an iterative algorithm based on geometric distance and local feature weighting named weighted correlation coefficient iterative closest point algorithm (WCC_ICP). The algorithm first establishes the corresponding relationship of points using linear representation and constructs an optimization model with constraints. Subsequently, the constrained optimization model is converted into an unconstrained optimization model by Lagrangian multiplication. Finally, an iterative technique is used to solve the optimal rigid body transformation of the optimized model. In addition, in each iteration, the identification ability of the local feature operator of points is used to repair the corresponding relation of points, which reduces the risk of the iterative algorithm falling into the local minima. Numerical experiments show that the proposed WCC_ICP algorithm reduces the risk of the iterative algorithm falling into local minima.

Accurate Covariance Estimation for Pose Data From Iterative Closest Point Algorithm

<h3>Abstract</h3> One of the fundamental problems of robotics and navigation is the estimation of the relative pose of an external object with respect to the observer. A common method for computing the relative pose is the iterative closest point (ICP) algorithm, where a reference point cloud of a known object is registered against a sensed point cloud to determine relative pose. To use this computed pose information in downstream processing algorithms, it is necessary to estimate the uncertainty of the ICP output, typically represented as a covariance matrix. In this paper, a novel method for estimating uncertainty from sensed data is introduced.

Easy to Calibrate: Marker-Less Calibration of Multiview Azure Kinect

Reconstructing a three-dimensional (3D) environment is an indispensable technique to make augmented reality and augmented virtuality feasible. A Kinect device is an efficient tool for reconstructing 3D environments, and using multiple Kinect devices enables the enhancement of reconstruction density and expansion of virtual spaces. To employ multiple devices simultaneously, Kinect devices need to be calibrated with respect to each other. There are several schemes available that calibrate 3D images generated from multiple Kinect devices, including the marker detection method. In this study, we introduce a markerless calibration technique for Azure Kinect devices that avoids the drawbacks of marker detection, which directly affects calibration accuracy; it offers superior user-friendliness, efficiency, and accuracy. Further, we applied a joint tracking algorithm to approximate the calibration. Traditional methods require the information of multiple joints for calibration; however, Azure Kinect, the latest version of Kinect, requires the information of only one joint. The obtained result was further refined using the iterative closest point algorithm. We conducted several experimental tests that confirmed the enhanced efficiency and accuracy of the proposed method for multiple Kinect devices when compared to the conventional marker-based calibration.

Расчет параметров наземного лазерного сканирования промышленных объектов

Terrestrial laser scanning of industrial facilities has a number of features, among which are: a large number of laser scanner installation stations, short distances between stations, complex geometry and differences in the reflectivity of the measurement objects materials, as well as a large amount of data derived from the work. To achieve the required accuracy in solving different tasks these features must be taken into account when performing measurements, as well as when processing measurement results. Based on the analysis early research and practical experience, authors propose a generalized technological scheme for terrestrial laser scanning of industrial facilities. In the article the scan settings also calculated, including setting of scan resolution, the choice of the distance to the object and the permissible beam incidence angle. Proposals are also given for the point cloud registration and datum transformation using an Iterative Closest Point Algorithm. The performed studies can be useful for precalculating the accuracy of the terrestrial laser scanning results.

Coarse Point Cloud Registration Based on Variational Functionals

Point cloud collection forming a 3D scene typically uses information from multiple data scans. The common approach is to register the point cloud pairs consequentially using a variant of the iterative closest point (ICP) algorithm, but most versions of the ICP algorithm only work correctly for a small movement between two point clouds. This makes it difficult to accumulate multiple scans. Global registration algorithms are also known, which theoretically process point clouds at arbitrary initial positions. Recently, a multiparameter variational functional was described and used in the ICP variant to register point clouds at arbitrary initial positions. The disadvantage of this algorithm was the need for manual selection of parameters. In this paper, a modified version of the algorithm with automatic selection of the model parameters is proposed. The proposed algorithm is a fusion of the ICP and RANSAC concepts. Moreover, the algorithm can be parallelized. The performance of the proposed algorithm is compared with that of known global registration algorithms.

Three‐dimensional point cloud reconstruction of inverse synthetic aperture radar image sequences based on back projection and iterative closest point fusion

AbstractIn order to recover the three‐dimensional (3D) structure of the target from sequential inverse synthetic aperture radar (ISAR) images, the factorisation method is generally used. It requires a large number of high‐quality matched feature points from different ISAR images. If the number of extracted feature points is insufficient, the restored 3D structure is not obvious. Furthermore, the mismatching of feature points will greatly affect the quality of target reconstruction. However, the factorisation method only uses the information from the ISAR images, while that from imaging geometry is not sufficiently considered. ISAR imaging is a kind of projection, and the projection plane information could be taken into account for the 3D reconstruction. Hence, a new 3D reconstruction method for stable targets from sequential ISAR images is proposed in this paper. Firstly, the ISAR images are preprocessed with the CLEAN algorithm. The maximum between‐cluster variance method is applied to extract feature points from the processed images. Moreover, the image projection plane and projection equation corresponding to different ISAR images are analysed by using imaging geometry information. According to the projection equation, the feature points are back‐projected (BP) to the 3D space. Finally, the 3D point clouds obtained by the BP from multiple radar images are fused by the iterative closest point algorithm to restore the 3D structure of the target. The simulation and experiment results show the effectiveness and robustness of this method.

Solution of The Variational Registration Problem Based on Iterative Closest Point Algorithms for Various Types of Geometric Transformations

Solution of The Variational Registration Problem Based on Iterative Closest Point Algorithms for Various Types of Geometric Transformations

A calculation method of phenotypic traits based on three-dimensional reconstruction of tomato canopy

Accurate and rapid acquisition of tomato canopies’ phenotypic traits was significant for variety breeding, cultivation, and scientific management. Manual measurements were time-consuming, laborious, and error-prone. The large devices in the field were lack of mobility, while the single perspective was limited by the environment’s obstruction, making it challenging to achieve high-throughput detection of tomato plants’ phenotypes. Therefore, a method for high-throughput detection of tomato canopy’s phenotypic traits was proposed based on three-dimensional (3D) structure reconstruction with multi-views. First, the tomato variety named Dongnong 708 was set as an experimental object. The acquisition platform was constructed using three Kinect 2.0 sensors to acquire a full range point cloud of tomato canopy at crucial growth stages, including the initial flowering stage, florescence, and primary fruit stage. Second, the background and interference noises were removed by Conditional filtering and Statistical Outlier Removal (SOR) filtering. Then, its structural characteristic points were extracted, and spatial position was registered by combining with Intrinsic Shape Signatures (ISS) and Iterative Closest Point (ICP) algorithms. In addition, the 3D-Normal Distributions Transform (NDT) algorithm was used to realize accurate registration of three perspectives’ point clouds. Compared with NDT and ICP algorithms, the average error of the proposed methods here was 0.027, reduced by 0.02 and 0.04, respectively. Finally, the contour of tomato canopy was extracted using the AlphaShape algorithm. Based on the reconstructed result, plant height, canopy width, and leafstalk angle, were calculated. The results showed that the correlation coefficients were 0.9615, 0.809 and 0.9014 between the calculated values and measured values. The average errors were 1.38 cm, 5.1° and 1.92 cm, respectively. The methods proposed in the paper could be used as a rapid detection method for quantitative indexes of the phenotypic traits for the tomato canopy and provide technical support for breeding, scientific cultivation, and environmental regulation.

Registration based assessment of femoral torsion for rotational osteotomies based on the contralateral anatomy

BackgroundComputer-assisted techniques for surgical treatment of femoral deformities have become increasingly important. In state-of-the-art 3D deformity assessments, the contralateral side is used as template for correction as it commonly represents normal anatomy. Contributing to this, an iterative closest point (ICP) algorithm is used for registration. However, the anatomical sections of the femur with idiosyncratic features, which allow for a consistent deformity assessment with ICP algorithms being unknown. Furthermore, if there is a side-to-side difference, this is not considered in error quantification.The aim of this study was to analyze the influence and value of the different sections of the femur in 3D assessment of femoral deformities based on the contralateral anatomy.Material and methods3D triangular surface models were created from CT of 100 paired femurs (50 cadavers) without pathological anatomy. The femurs were divided into sections of eponymous anatomy of a predefined percentage of the whole femoral length. A surface registration algorithm was applied to superimpose the ipsilateral on the contralateral side. We evaluated 3D femoral contralateral registration (FCR) errors, defined as difference in 3D rotation of the respective femoral section before and after registration to the contralateral side. To compare this method, we quantified the landmark-based femoral torsion (LB FT). This was defined as the intra-individual difference in overall femoral torsion using with a landmark-based method.ResultsContralateral rotational deviation ranged from 0° to 9.3° of the assessed femoral sections, depending on the section. Among the sections, the FCR error using the proximal diaphyseal area for registration was larger than any other sectional error. A combination of the lesser trochanter and the proximal diaphyseal area showed the smallest error. The LB FT error was significantly larger than any sectional error (p < 0.001).ConclusionWe demonstrated that if the contralateral femur is used as reconstruction template, the built-in errors with the registration-based approach are smaller than the intraindividual difference of the femoral torsion between both sides. The errors are depending on the section and their idiosyncratic features used for registration. For rotational osteotomies a combination of the lesser trochanter and the proximal diaphyseal area sections seems to allow for a reconstruction with a minimal error.