Improved efficiency of road sign detection and recognition by employing Kalman filter

Citation

Zakir U, Hussain A, Ali L & Luo B (2013) Improved efficiency of road sign detection and recognition by employing Kalman filter. In: Liu D, Alippi C, Zhao D & Hussain A (eds.) Advances in Brain Inspired Cognitive Systems: 6th International Conference, BICS 2013, Beijing, China, June 9-11, 2013. Proceedings. Lecture Notes in Computer Science, 7888. 6th International Conference on Brain Inspired Cognitive Systems, BICS 2013, Beijing, China, 09.06.2013-11.06.2013. Berlin Heidelberg: Springer, pp. 216-224. http://link.springer.com/chapter/10.1007/978-3-642-38786-9_25#; https://doi.org/10.1007/978-3-642-38786-9_25

Abstract
This paper describes an efficient approach towards road sign detection, and recognition. The proposed system is divided into three sections namely: Road Sign Detection where Colour Segmentation of the road traffic signs is carried out using HSV colour space considering varying lighting conditions and Shape Classification is achieved by using Contourlet Transform, considering possible occlusion and rotation of the candidate signs. Road Sign Tracking is introduced by using Kalman Filter where object of interest is tracked until it appears in the scene. Finally, Road Sign Recognition is carried out on successfully detected and tracked road sign by using features of a Local Energy based Shape Histogram (LESH). Experiments are carried out on 15 distinctive classes of road signs to justify that the algorithm described in this paper is robust enough to detect, track and recognize road signs under varying weather, occlusion, rotation and scaling conditions using video stream.

Keywords
Road Signs; HSV; Contourlet Transform; LESH; Colour Segmentation; Autonomous Vehicles; Kalman Filter; SVM