Abstract
Recently, the Coherent Point Drift (CPD) algorithm has become a very popular and efficient method for point set registration. However, this method does not take into consideration the neighborhood structure information of points to find the correspondence and requires a manual assignment of the outlier ratio. Therefore, CPD is not robust for large degrees of degradation. In this paper, an improved method is proposed to overcome the two limitations of CPD. A structure descriptor, such as shape context, is used to perform the auxiliary calculation of the correspondence, and the proportion of each GMM component is adjusted by the similarity. The outlier ratio is formulated in the EM framework so that it can be automatically calculated and optimized iteratively. The experimental results on both synthetic data and real data demonstrate that the proposed method described here is more robust to deformation, noise, occlusion, and outliers than CPD and other state-of-the-art algorithms.
Highlights
Point set registration is one of the main methods for image registration
The goal of the point set registration is to align the model point set onto the scene point set, where the model point set is presented by blue pluses and the scene point set is red circles
For the purpose of evaluating the performance of the proposed method, the results of our method are compared with four state-of-the-art algorithms: GMMREG [10], TPS-RPM [8], RPM-L2E [27], and Coherent Point Drift (CPD) [9]
Summary
Point set registration is one of the main methods for image registration. As a fundamental component of the computer vision field, point set registration is often used in medical image processing [1,2,3], pattern recognition [4], and remote sensing image processing [5, 6]. Point set registration is divided into either rigid or non-rigid registration. Rigid registration is a relatively simple process that mainly processes the scaling, translation, and rotation of the point sets. Generalized rigid registration includes affine and projection transformation. For non-rigid registration, the transformation form is non-rigid, and it is difficult to accurately determine the transformation model, especially if there is a large degree of degradation, such as deformation, occlusion, noise, or outliers
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
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 callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
