Date of Publication :7th May 2016
Abstract: Abundance in text along with development of digitalized camera system lead to a new era of recognition and analysis. Traffic sign board which meant for road safety is all time a matter of fact for the traffic safety. The traffic board which carries signs along with informative texts are to be recognized for safe driving. Apart from normal sign board recognition, here the informative boards are also considered. This is a fewer step towards driver assistive system. The scene recognition faces a lot of challenges like motion blur, occlusion and lighting factors. The goal is to implement a system (model) for recognizing the text and signs in the traffic board and to create sound alert. The HOG based vector machine (SVM) classifies the traffic sign and the text is recognized using MSER based OCR. Advantages of the proposed system are well condition under illumination changes sound alert and text recognition is an added advantage with other methods.
Reference :
-
[1] A. Møgelmose, M. M. Trivedi. and T. B. Moeslund. Vision-based traffic sign detection and analysis for intelligent driver assistance systems: Perspective sand survey. IEEE Trans. Intell. Transp. Syst., vol. 13,no. 4, pp. 1484–1497, December2012
[2] F. Zaklouta. and B. Stanciulescu.. Warning traffic sign recognition using a HOG-based K-d tree.in Proc. IEEE IV Symp. pp. 1019–1024.June 2011
[3] J. Greenhalgh, M. Mirmehdi. and S. Member.. Realtime detection and recognition of road traffic signs. IEEE Trans. Intell. Transp. Syst. Vol. 13, no. 4,pp. 1498–1506, December. 2012
[4] K. Elagouni, C. Garcia, F. Mamalet, and P. Se´billot, “Combining Multi-Scale Character Recognition and Linguistic Knowledge for Natural Scene Text OCR,” Proc. Int’l Workshop Document Analysis Systems, pp. 120-124, Mar. 2012
[5] M. Boumediene, C. Cudel, M. Basset. and A. Ouamri. Triangular traffic signs detection based on RSLD algorithm. Mach. Vis. Appl. Vol. 24, no. 8, pp. 1721– 1732, November. 2013
[6] Yamazoe, M. Etoh, T. Yoshimura, and K. Tsujino, “Towards integerated Scene Text Reading,” Proc. Int’l Conf. Document Analysis and Recognition, pp. 359- 363, dec.2014