Lane Characterization Under Challenging Scenarios For Autonomous Driving System

Authors

  • Upendra Suddamalla Embedded Innovation Lab, EIS, Tata Consultancy Services, Bangalore, India
  • Varsha Singal Embedded Innovation Lab, EIS, Tata Consultancy Services, Bangalore, India
  • Susmitha Mohan Embedded Innovation Lab, EIS, Tata Consultancy Services, Bangalore, India

DOI:

https://doi.org/10.14738/aivp.46.2538

Keywords:

Lane Detection, Lane Color, Lane Structure, Relative Color, Zero Crossing, Driver Assistance, Autonomous driving

Abstract

Advanced Driver Assistance systems (ADAS) have seen increasing popularity due to their importance in automotive driver safety and autonomous driving. These systems analyze data from several sensors mounted on the vehicle to detect lanes, obstacles and traffic conditions to ensure safe driving. Lane markings on road with different color and structure provide information on safe drive zone and other traffic restrictions on road. Typical ADAS solutions depend on vision based sensors for lane detection. Here we propose an efficient algorithm for detecting type and color of the lane marks as this information plays critical role in taking the decision for safety features such as lane change and lane keep assist. Our algorithm is pluggable to any state-of-art lane detection algorithm and provides lane type and color for straight, curvy roads. The proposed method is tested on various challenging scenarios and results are promising.

References

(1) ISO 17361 Intelligent transport systems - Lane departure warning systems Performance requirement and test procedures, First edition 2007.

(2) G Kaur, D Kumar: “Lane Detection Techniques: A Review,” International Journal of Advanced Computer Science and Applications, vol.112, no.10, pp.0975 8887, February 2015.

(3) U Suddamalla, S Kundu, S Farkade: “A Novel Algorithm of Lane Detection Addressing Varied Scenarios of Curved and Dashed Lanemarks,” International Conference on Image Processing Theory, Tools and Applications, pp.87-92, IPTA 2015.

(4) JM Collado, C Hilario, A De La Escalera: “Adaptative Road Lanes Detection and Classification,”

(5) CH Rodrguez-Garavito, A Ponz, F Garca: “COMPUTER VISION APPLIED TO ROAD LINES RECOGNITION USING MACHINE LEARNING,” The 7th International Conference on Information Technology, ICIT 2015.

(6) MB de Paula, CR Jung: “Real-time detection and classification of road lane markings,” 2013 XXVI Conference on Graphics, Patterns and Images, pp.1530- 1834, IEEE 2013.

(7) S Vacek, C Schimmel, R Dillmann: “Road-marking analysis for autonomous vehicle guidance,”

(8) S Suchitra, RK Satzoda: “Identifying Lane Types: A Modular Approach,” Proceedings of the 16th International IEEE Annual Conference on Intelligent Transportation Systems, (ITSC 2013), 2013 IEEE.

(9) LNP Boggavarapu, RS Vaddi, KR Anne: “A robust multi color lane marking detection approach for Indian scenario,” International Journal of Advanced Computer Science and Applications, vol.2, no.5, 2011.

(10) TT Tran, CS Bae, YN Kim, HM Cho, SB Cho: “An Adaptive Method for Lane Marking Detection Based on HSI Color Model,” International Journal of Advanced Computer Science and Applications, ICIC 2010, CCIS 93,pp.304-311,2010.

(11) HC Choi, SY Oh: “Illumination Invariant Lane Color Recognition by using Road Color Reference and Neural Networks,” International Journal of Advanced Computer Science and Applications, 2010 IEEE.

(12) H Lee, S Park, K Choi: “Support Vector Machines For Understanding Lane Color and Sidewalks,” International Journal of Computer, Electrical, Automation, Control and Information Engineering, vol.3, no.2, 2009.

(13) http://vision.caltech.edu/malaa/datasets/caltech-lanes/

(14) http://www.istockphoto.com/in/stock-photos

Downloads

Published

2017-01-08

How to Cite

Suddamalla, U., Singal, V., & Mohan, S. (2017). Lane Characterization Under Challenging Scenarios For Autonomous Driving System. European Journal of Applied Sciences, 4(6), 18. https://doi.org/10.14738/aivp.46.2538