Use of Computer Vision for White Line Detection for Robotic Applications

Image processing and its use for object detection that allows to alter decisions based on what is being observed remains a challenge today. As participants in the Intelligent Ground Vehicle Competition (IGVC) 2018 and 2019, we were tasked to build an autonomous vehicle that is capable of maneuvering and traversing through a grassy course delineated with white lines and laid out with scattered obstacles. In order to remain within the white line boundaries, computer vision was used with the implementation of OpenCV and Python for image processing and white line detection. To accomplish the white line detection task, various image processing techniques such as change of color space, image filtering, blurring, histogram equalizing, and edge detection are utilized. Testing of the image processing techniques was performed using prerecorded videos from previous competitions as well as videos taken using the autonomous vehicle’s ZED camera. In both test cases there were challenges due to differences in lighting conditions, image clarity between cameras used for both video types, and the angle of the camera.

[1]  Raja Chatila,et al.  Performance measure for the evaluation of mobile robot autonomy , 2006, Proceedings 2006 IEEE International Conference on Robotics and Automation, 2006. ICRA 2006..

[2]  Henryk Palus,et al.  Representations of colour images in different colour spaces , 1998 .

[3]  Joseph F. Coughlin,et al.  In-vehicle technology for self-driving cars: Advantages and challenges for aging drivers , 2014 .

[4]  Yann LeCun,et al.  Off-Road Obstacle Avoidance through End-to-End Learning , 2005, NIPS.

[5]  Steffen Junginger,et al.  Mobile Robot for Life Science Automation , 2013 .

[6]  Lizhuang Ma,et al.  Comparison of different color spaces for image segmentation using graph-cut , 2015, 2014 International Conference on Computer Vision Theory and Applications (VISAPP).

[7]  Sergey Levine,et al.  Deep spatial autoencoders for visuomotor learning , 2015, 2016 IEEE International Conference on Robotics and Automation (ICRA).

[8]  Srikanth Saripalli,et al.  Road detection from aerial imagery , 2012, 2012 IEEE International Conference on Robotics and Automation.

[9]  Anthony Stentz,et al.  First results in vision-based crop line tracking , 1996, Proceedings of IEEE International Conference on Robotics and Automation.

[10]  Joan Serrat,et al.  Robust lane markings detection and road geometry computation , 2010 .

[11]  Tapas Kumar,et al.  Comparative Analysis of Edge Detection between Gray Scale and Color Image , 2016 .

[12]  Dana H. Ballard,et al.  Generalizing the Hough transform to detect arbitrary shapes , 1981, Pattern Recognit..

[13]  Thomas S. Huang,et al.  Computer Vision: Evolution And Promise , 1996 .

[14]  Mohamed Aly,et al.  Real time detection of lane markers in urban streets , 2008, 2008 IEEE Intelligent Vehicles Symposium.

[15]  John F. Canny,et al.  A Computational Approach to Edge Detection , 1986, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[16]  Kok Kiong Tan,et al.  Vision-based approach towards lane line detection and vehicle localization , 2015, Machine Vision and Applications.

[17]  Jenay M. Beer,et al.  Toward a framework for levels of robot autonomy in human-robot interaction , 2014, Journal of human-robot interaction.

[18]  Rafael Grompone von Gioi,et al.  LSD: A Fast Line Segment Detector with a False Detection Control , 2010, IEEE Transactions on Pattern Analysis and Machine Intelligence.