Multiple point sets registration based on Expectation Maximization algorithm. (August 2018)

• Record Type: Journal Article
• Authors: Zhou, Yiqiong; Xu, Siyu; Jin, Congcong; Guo, Ziyi
• Is Part Of: Computers & electrical engineering. Volume 70(2018)
• Page Start: 1
• Page End: 11
• Publication Date: 2018-08
• Subjects: Multi-view registration -- Noise -- Gaussian distribution -- Expectation Maximization algorithm -- Iterative Closest Point; Computer engineering -- Periodicals; Electrical engineering -- Periodicals; Electrical engineering -- Data processing -- Periodicals; Ordinateurs -- Conception et construction -- Périodiques; Électrotechnique -- Périodiques; Électrotechnique -- Informatique -- Périodiques; Computer engineering; Electrical engineering; Electrical engineering -- Data processing; Periodicals; Electronic journals; 621.302854
• DOI: 10.1016/j.compeleceng.2018.05.026
• Languages: English
• ISSNs: 0045-7906

Abstract

Abstract: Many methods have been proposed and improved to deal with multi-view point cloud registration. Most of them are based on the classical method, namely Iterative Closest Point(ICP), which is fast and accurate in most cases. However in the presence of noise and outliers, the equal weights ICP assigned to all correspondences would lead to unsatisfactory registration results. To address this issue, this paper proposes a new automatic multi-view registration method based on Expectation Maximization(EM). A Gaussian distribution on representing the relationship between point-sets according to their distance was introduced first. Then EM is brought in to optimize the likelihood function formulated with above Gaussian distribution. Through an iterative method the multi-view registration results can be obtained at last. The experimental results demonstrate accuracy and robustness of our methods over four state-of-the-art algorithms, especially when noisy data exist.