Obstacle detection for intelligent vehicles using semi-supervised active learning

Wang, Liantao, Hu, Xuelei and Du, Pei (2012). Obstacle detection for intelligent vehicles using semi-supervised active learning. In: International Conference on Automatic Control and Artificial Intelligence (ACAI 2012). International Conference on Automatic Control and Artificial Intelligence (ACAI 2012), Xiamen, China, (1459-1462). 3-5 March 2012. doi:10.1049/cp.2012.1256


Author Wang, Liantao
Hu, Xuelei
Du, Pei
Title of paper Obstacle detection for intelligent vehicles using semi-supervised active learning
Conference name International Conference on Automatic Control and Artificial Intelligence (ACAI 2012)
Conference location Xiamen, China
Conference dates 3-5 March 2012
Proceedings title International Conference on Automatic Control and Artificial Intelligence (ACAI 2012)
Place of Publication Piscataway, NJ, United States
Publisher IEEE
Publication Year 2012
Sub-type Fully published paper
DOI 10.1049/cp.2012.1256
ISBN 9781849195379
Start page 1459
End page 1462
Total pages 4
Collection year 2013
Language eng
Abstract/Summary Reliable environment perception system is critical to path planning and autonomous navigation of intelligent vehicles. One feasible way to percept environment is obstacle detection by classifying image patches as obstacle or non-obstacle. Accurate classification system depends on appropriate training data. For intelligent vehicles, a large number of images can be easily obtained while labeling them is tedious. Additionally, the accuracy is limited for the scene diversity. In this paper, we propose a semi-supervised active learning algorithm which can exploit the most certain unlabeled examples and query the most informative examples to enhance the performance of classifiers. In view of the scene diversity, we present a two-level classification system which first distinguishes the scene category using level-I classifier before calling the suitable level-II classifier to detect obstacles. The experimental results demonstrate the efficiency of our algorithm and two-level classification system.
Keyword Obstacle detection
Intelligent vehicle
Semisupervised learning
Active learning
Q-Index Code E1
Q-Index Status Provisional Code
Institutional Status Non-UQ

 
Versions
Version Filter Type
Citation counts: Scopus Citation Count Cited 0 times in Scopus Article
Google Scholar Search Google Scholar
Created: Mon, 16 Dec 2013, 12:40:24 EST by Xuelei Hu on behalf of School of Information Technol and Elec Engineering