Hits:

Indexed by:期刊论文

Date of Publication:2011-08-01

Journal:ICIC Express Letters

Included Journals:EI、Scopus

Volume:5

Issue:8 B

Page Number:2741-2746

ISSN No.:1881803X

Abstract:In this paper, we consider the problem of 3D point cloud classification for ground robots in outdoor environment. This problem is important for various tasks from environment modeling to obstacle avoidance for autonomous mobile robot navigation. In order to improve the description of natural scene, we propose novel features with the consideration of surface condition and environmental characters. By introducing a fuzzy inference system, feature fusion is implemented to produce a high-level feature which represents the inherent fuzziness of natural objects. The classification of 3D point cloud is approached by a fuzzy entropy method in the local area. Experimental results demonstrate the validity of the proposed approaches. ? 2011 ISSN 1881-803X.