Abstract
The iterative closest point (ICP) algorithm is widely used in three-dimensional (3D) point cloud registration, and it is very stable and robust. However, its biggest drawback is being easily trapped in a local optimal solution, which results in the incorrect registration result. Currently, there is neither a clear effective range to define whether the ICP algorithm will fall into a local optimum nor a study providing a comprehensive evaluation of the ICP algorithm. In this paper, we take the overlap ratio, angle, distance, and noise as the influencing factors of ICP and evaluate the validity, robustness, accuracy, and efficiency of point-to-point and point-to-plane ICP by using four datasets. We first analyze the effective ranges of the two ICP algorithms with respect to overlap ratio, angle, and distance and then propose a universal effective range for the three factors. Next, the effect of Gaussian noise on the validity and accuracy of two ICPs is evaluated. We also analyze the factors influencing ICP accuracy and explain their changing rules. We finally study the effect of different parameters on the efficiency. All results are compared by using point-to-point and point-to-plane ICP algorithms. The results show that the overlap ratio has no effect on validity, but it has a significant influence on accuracy. The angle has a great impact on the validity and efficiency of ICP but has no effect on accuracy. The distance only affects validity, has a limited effect on efficiency and no effect on accuracy. Meanwhile, Gaussian noise has a little effect on the validity. In addition, the general effective range of point-to-point ICP is larger than that of point-to-plane ICP, but the point-to-plane ICP algorithm presents a better efficiency. The point-to-point ICP is more robust to Gaussian noise with respect to validity, while the point-to-plane is more resistant in terms of accuracy.
Highlights
The iterative closest point (ICP) algorithm [1] has long been regarded as the most classic algorithm in point cloud data registration, and it has been widely used for many years
VALIDITY OF POINT-TO-POINT ICP 1) VALIDITY BASED ON SAMPLING ANGLE - ROTATION ANGLE To verify the registration results of two datasets with different sampling angles and rotation angles, we uniformly use the dataset with 0-degree sampling angle and 0-degree rotation angle as the target dataset
In this study, the ICP algorithms in point cloud registration are evaluated
Summary
The ICP algorithm [1] has long been regarded as the most classic algorithm in point cloud data registration, and it has been widely used for many years. The shortcomings are obvious, for example, the computational efficiency is too low, the overlap ratio of the two datasets is high, the convergence domain is narrow, and it is easy to fall into a local optimal solution [2]. Working on perfecting ICP, including improving the efficiency of the ICP algorithm, increasing the accuracy of the algorithm and solving the problem of local optimal solutions. Improving the efficiency of the ICP algorithm is mainly achieved by reducing the number of points participating in the operation [3], increasing the search speed of the nearest neighbor [4], improving the reliability of the established point correspondence [5], registering multiview point sets [6], and reducing the number of iterations [7], which has achieved relatively obvious results. After each iteration of ICP, the distance between the nearest neighbor in the two datasets is calculated so that the point-to-point correspondence can
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
