Research of lane detection and recognition technology based on morphology feature

With the increasing of vehicle, people pay much attention to Intelligent Vehicle Visual Navigation System. Lane detection is the most important function of Intelligent Vehicle Visual Navigation System. When selected seed points are correct, the accuracy of the method of road region extraction based on regional growth is high. This method can identify lane region exactly. But when selected seed points are wrong, lane identification will fail, and lead into some interference information. In this paper, the deficiency of this method is improved. It uses this method to identify lane feature region, and introduces the area threshold value to filter scanned region growing area. This can reduce the interference of useless information on lane identification.

[1]  David A. Clausi,et al.  IRGS: Image Segmentation Using Edge Penalties and Region Growing , 2008, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[2]  Zequn Guan,et al.  An Improved Region Growing Algorithm for Image Segmentation , 2008, 2008 International Conference on Computer Science and Software Engineering.

[3]  Zehang Sun,et al.  Quantized wavelet features and support vector machines for on-road vehicle detection , 2002, 7th International Conference on Control, Automation, Robotics and Vision, 2002. ICARCV 2002..

[4]  Yoshiki Kobayashi,et al.  Multitype lane markers recognition using local edge direction , 2002, Intelligent Vehicle Symposium, 2002. IEEE.

[5]  Sunghoon Kim,et al.  A Framework for the Development of Robust Lane Recognition Systems , 2007, 2007 IEEE Intelligent Transportation Systems Conference.

[6]  Guangming Xiong,et al.  High Speed Lane Recognition under Complex Road Conditions , 2008, 2008 IEEE Intelligent Vehicles Symposium.