DocumentCode :
627237
Title :
Night mode prohibitory traffic signs detection
Author :
Biswas, Rubel ; Khan, Ajmal ; Alom, Md Zahangir ; Khan, Mahrukh
Author_Institution :
Dept. of Comput. Sci. & Eng., BRAC Univ., Dhaka, Bangladesh
fYear :
2013
fDate :
17-18 May 2013
Firstpage :
1
Lastpage :
5
Abstract :
Prohibitory traffic signs play an important role in guiding, warning and regulating traffic system. As driving over the speed limit is often the major cause of accidents, detecting this group of prohibitory signs may reduce this danger. This paper presents an approach to detecting speed limit signs at night mode which is based on Multi-Scale Retinex Color Restoration and Hough Transform. Experiment to check the strength of this approach shows that approximately 96.6% of the prohibitory traffic signs invoked for this test were successfully detected. This test was carried out at dark mode images from different country.
Keywords :
Hough transforms; image colour analysis; image restoration; object detection; traffic engineering computing; Hough transform; dark mode images; multiscale retinex color restoration; night mode prohibitory traffic sign detection; speed limit sign detection; traffic system guidance; traffic system regulation; traffic system warning; Image color analysis; Image edge detection; Image restoration; Intelligent vehicles; Roads; Transforms; Vehicles; Hough transform; MSRCR; Traffic signs;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Informatics, Electronics & Vision (ICIEV), 2013 International Conference on
Conference_Location :
Dhaka
Print_ISBN :
978-1-4799-0397-9
Type :
conf
DOI :
10.1109/ICIEV.2013.6572589
Filename :
6572589
Link To Document :
بازگشت