Available Lane Detection Based on Radon Transform

Abstract:

To solve lane detection problem in the system of autonomous vehicle, this paper proposes a method of texture segment based on perspective transformation. In this paper, firstly road images were captured through cameras installed on the vehicle, then make a perspective transform to road plane, so that the road and the non-road texture information effectively stand out. After calculation of the texture trend in the transformed image, radon transform can effectively distinguish between the road and the non-road area, and achieve the purpose of the texture regional segment. Experiments prove that this method can be used on the lane detection, which eliminate barriers and road borders effectively.

• Periodical: Advanced Materials Research (Volume 1046)
• Pages: 415-424
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.1046.415
• Online since: October 2014
• Authors: Xi Xi Deng, Xiao Nian Wang*, Jin Zhu
• Keywords: Autonomous Vehicle, Lane Detection, Perspective Transform, Radon Transform, Texture Direction

References

[1] Bucher T, Curio C, et al. Image processing and behavior planning for intelligent vehicles [J]. IEEE Transactions on Industrial Electronics, 2003, 50(1).

[2] Yinghua He, Hong Wang, Bo Zhang. Color-Based Road Detection in Urban Traffic Scenes. IEEE Transactions on Intelligent Transportation Systems, Dec (2004).

DOI: 10.1109/itsc.2003.1252047

[3] N Mechat, N Saadia, N.K. M'Sirdi, Lane Detection and Tracking by Monocular Vision System in Road Vehicle. 5th International Congress on Image and Signal Processing (CISP 2012).

DOI: 10.1109/cisp.2012.6469683

[4] Rongben Wang, Feng You, Lane Detection for High Speed Intelligent Vehicle Based on Machine Vision, Computer Engineering and Applications, 2004, pp.18-21.

[5] Desong Ma．Computer Vision : Calculation theory and algorithm．Science Press，(1997).

[6] YuJin Zhang. Image Engineering. Beijing, Tsinghua University Press (1999).

[7] Tao Sun, Shuming Tang, Jinqiao Wang. A Robust Lane Detection Method for Autonomous Car -like Robot, 4th International Conference on Intelligent Control and Information Processing (ICICIP) . IEEE, (2013).

DOI: 10.1109/icicip.2013.6568100

[8] Youchun Xu．Research on Machine Vision Based Vehicle Active Safety Technology. Beijing: Department of Automotive Engineering. Tsinghua University, (2003).

[9] Chunzhao Guo, Seiichi Mita. Drivable Road Region Detection based on Homography Estimation with Road Appearance and Driving State Models. Toyota Technological Institute. International Conference on Autonomous Robots and Agents, Feb, 2009 IEEE.

DOI: 10.1109/icara.2000.4803961

[10] Lei Guo, Ying Ma, Inverse perspective transform based on directional 2-D interpolation, Journal Of Tsinghua University (Sci & Tech) (2006).

[11] Ziqiong Liu, Shengjin Wang. ROI Perspective Transform Based Road Marking Detection and Recognition, Tsinghua University, Beijing, China, pp.844-846, ICALIP 2012 IEEE.

[12] J. K Kourosh, S.Z. Hamid, Rotation-Invariant Mutire solution Texture Analysis Using Radon and Wavelet Transform" IEEE Trans. Image Process, vol14, pp.783-793, June (2005).

DOI: 10.1109/tip.2005.847302

[13] J. h, Li, Q. Pan. Image recognition using Radon transform. Intelligent Transportation Systems， vol. 1, pp: 741-744，Jan (2003).

[14] Jianqing Shen，Detection Method for Fabric Texture Direction Based on Wavelet Transform, Computer Engineering, Vol. 33, No. 6, pp.182-184, (2007).

[15] Shin H S, Lee C, Lee M. Adaptive threshold method for the peak detection of photoplethysmographic waveform[J]. Computers in Biology and Medicine, pp, 1145-1152, (2009).

DOI: 10.1016/j.compbiomed.2009.10.006