A terrain classification method for UGV autonomous navigation based on SURF

Author

Lee, Seung-Youn ; Kwak, Dong-Min

Author_Institution

Unmanned Ground Vehicle Technol. Directorate, Agency for Defense Dev., Daejeon, South Korea

fYear

2011

fDate

23-26 Nov. 2011

Firstpage

303

Lastpage

306

Abstract

The ability to navigate autonomously in off-road terrain is critical technology needed for unmanned ground vehicle (UGV). This paper presents a vision-based off-road terrain classification method that is robust despite environmental variation caused by weather changes. In order to cope with an overall image brightness variation, we use speeded-up robust features (SURF), and neural network classifier. Experimental results for real off-road images show that proposed method has a better performance than wavelet based one especially in case of large brightness variation.

Keywords

feature extraction; image classification; mobile robots; neural nets; robot vision; UGV autonomous navigation; brightness variation; environmental variation; neural network classifier; speeded-up robust features; unmanned ground vehicle; vision based off-road terrain classification method; wavelet; weather changes; Brightness; Classification algorithms; Feature extraction; Image color analysis; Robustness; Support vector machine classification; Training; SURF; UGV; autonomous navigation; terrain classification; wavelet features;

fLanguage

English

Publisher

ieee

Conference_Titel

Ubiquitous Robots and Ambient Intelligence (URAI), 2011 8th International Conference on

Conference_Location

Incheon

Print_ISBN

978-1-4577-0722-3

Type

conf

DOI

10.1109/URAI.2011.6145981

Filename

6145981