Int. J. of Wireless and Mobile Computing   »   2016 Vol.11, No.3

 

 

Title: A novel method of natural landmark extraction and description for vSLAM

 

Authors: Quan-De Yuan; Yi Guan; Bing-Rong Hong; Song-Hao Piao; Ze-Su Cai

 

Addresses:
School of Computer Science and Technology, Harbin Institute of Technology, Harbin 150001, China; School of Electrical Engineering and Information Technology, Changchun Institute of Technology, Changchun 130012, China
School of Computer Science and Technology, Harbin Institute of Technology, Harbin 150001, China
School of Computer Science and Technology, Harbin Institute of Technology, Harbin 150001, China
School of Computer Science and Technology, Harbin Institute of Technology, Harbin 150001, China
School of Computer Science and Technology, Harbin Institute of Technology, Harbin 150001, China

 

Abstract: Landmark extraction and matching is a key part of vSLAM. In this paper, a method of landmark extraction and its description based on three-dimensional (3D) information of feature points is proposed. The robot obtains images from its environment via binocular vision, building 3D information of feature points under left camera coordinate system. Then, it uses the method based on improved mean shift algorithm to extract natural landmarks and construct its descriptor, which can be used in fast matching of two landmarks. This method can extract natural landmarks from unstructured environments and tolerate relatively low accuracy of pose estimation. It is suitable for navigation system of a mobile robot running in a complex environment. vSLAM experiments showed its effectiveness.

 

Keywords: mobile robots; natural landmarks; clustering; vSLAM; landmark extraction; landmark description; robot vision; binocular vision; mean shift algorithm; robot navigation.

 

DOI: 10.1504/IJWMC.2016.10002150

 

Int. J. of Wireless and Mobile Computing, 2016 Vol.11, No.3, pp.214 - 221

 

Submission date: 08 Jun 2016
Date of acceptance: 28 Sep 2016
Available online: 24 Dec 2016

 

 

Editors Full text accessAccess for SubscribersPurchase this articleComment on this article