A Self-Supervised Road Detection Method Based on Gabor Filter

Abstract:

Visual navigation is an important technology for Autonomous Guided Vehicles and road detection is the essential prerequisite. It is a challenging problem to distinguish the road or its boundaries in unstructured environments for the lack of discernible features. A self-supervised road detection method is proposed in this paper. Vanishing point is first detected using only four Gabor filters to precisely estimate the local dominant orientation at each pixel location, and an adaptive soft voting scheme is used to prevent tending to favor points that are high in the image. Training area is defined with vanishing point, which color feature is used for building self-supervised learning models to describe the road segment. Road patches are selected by measuring the roadness score. Experimental results have shown the efficiency of the method in terms of detection result and time saving.

• Periodical: Advanced Engineering Forum (Volumes 6-7)
• Pages: 763-767
• DOI: https://doi.org/10.4028/www.scientific.net/AEF.6-7.763
• Online since: September 2012
• Authors: Hua Yao Chang, Jun Zheng Wang, Jing Li
• Keywords: Gabor Filter, Road Detection, Vanishing Point

References

[1] J. C. McCall, M. M. Trivedi. Video-Based Lane Estimation and Tracking for Driver Assistance: Survey, System, and Evaluation [J]. IEEE Trans. on Intelligent Transportation Systems, 2006, 7(1): 20-37.

DOI: 10.1109/tits.2006.869595

[2] J. Crisman, C. Thorpe. Unscarf, a color vision system for the detection of unstructured roads [C]. IEEE International Conference on Robotics and Automation, 1991, 2496– 2501.

DOI: 10.1109/robot.1991.132000

[3] A. Wedel, T. Pock, C. Zach, etc. An improved algorithm for tv-l1 optical flow [C]. Statistical and Geometrical Approaches to Visual Motion Analysis: International Dagstuhl Seminar, 2009, 23-45.

DOI: 10.1007/978-3-642-03061-1_2

[4] J. R. Claudio, K. R. Christian. Lane following and lane departure using a linear-parabolic model [J]. Image and vision computing, 2005, 23: 1192-1202.

DOI: 10.1016/j.imavis.2005.07.018

[5] C. Rasmussen. Grouping dominant orientations for illstructured road following [C]. Proc. IEEE Comp. Soc. Conf. on Computer Vision and Pattern Recogintion. 2004, 470-477.

DOI: 10.1109/cvpr.2004.1315069

[6] H. Kong, J. Audibert, J. Ponce. General road detection from a single image [J]. IEEE Trans. Image Process, 2010, 19(8): 2211-2220.

DOI: 10.1109/tip.2010.2045715

[7] M. Ondrej, P. Petr, Z. Ludek, etc. Robust detection of shady and highlighted roads for monocular camera based navigation of UGV [C]. IEEE International Conf. on Robotics and Automation, 2011, 64-71.

DOI: 10.1109/icra.2011.5979773

[8] M. Peyman, S. A. Janusz, W. S. Wijesoma. Fast vanishing-point detection in unstructured enviroments [J]. IEEE trans. on Image Processing, 2012, 21(1): 425-430.

DOI: 10.1109/tip.2011.2162422