{"title":"The Improvement of Iterative Closest Point with Edges of Projected Image","authors":"Chen Wang","doi":"10.1016/j.vrih.2022.09.001","DOIUrl":null,"url":null,"abstract":"<div><h3>Background</h3><p>There are many regular-shape objects in the artificial environment. It is difficult to distinguish the poses of these objects, when only geometric information is utilized. With the development of sensor technologies, we can utilize other information to solve this problem.</p></div><div><h3>Methods</h3><p>We propose an algorithm to register point clouds by integrating color information. The key idea of the algorithm is that we jointly optimize dense term and edge term. The dense term is built similarly to iterative closest point algorithm. In order to build the edge term, we extract the edges of the images obtained by projecting the point clouds. The edge term prevents the point clouds from sliding in registration. We utilize this loosely coupled method to fuse geometric and color information.</p></div><div><h3>Results</h3><p>The experiments demonstrate that edge image approach improves the precision and the algorithm is robust.</p></div>","PeriodicalId":33538,"journal":{"name":"Virtual Reality Intelligent Hardware","volume":"5 3","pages":"Pages 279-291"},"PeriodicalIF":0.0000,"publicationDate":"2023-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Virtual Reality Intelligent Hardware","FirstCategoryId":"1093","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S209657962200095X","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"Computer Science","Score":null,"Total":0}
引用次数: 0
Abstract
Background
There are many regular-shape objects in the artificial environment. It is difficult to distinguish the poses of these objects, when only geometric information is utilized. With the development of sensor technologies, we can utilize other information to solve this problem.
Methods
We propose an algorithm to register point clouds by integrating color information. The key idea of the algorithm is that we jointly optimize dense term and edge term. The dense term is built similarly to iterative closest point algorithm. In order to build the edge term, we extract the edges of the images obtained by projecting the point clouds. The edge term prevents the point clouds from sliding in registration. We utilize this loosely coupled method to fuse geometric and color information.
Results
The experiments demonstrate that edge image approach improves the precision and the algorithm is robust.