Title :
Traffic sign recognition by distorted template matching
Author :
Takada, R. ; Katto, J.
Author_Institution :
Dept. of Comput. Sci. & Eng., Waseda Univ., Tokyo, Japan
Abstract :
Traffic sign recognition contributes to the safety of drivers and people around the car. The system analyzes the road ahead images taken by on-vehicle camera, but blurred and distorted images make it difficult to recognize correct traffic sign. In this paper, we propose a method that deforms from ellipses to perfect circles and applies template matching with distorted sign templates.
Keywords :
image matching; transforms; SIFT features; distorted sign templates; distorted template matching; on-vehicle camera; traffic sign recognition; Accuracy; Cameras; Feature extraction; Image color analysis; Image recognition; Roads; Vehicles; SIFT features; template matching; traffic sign recognition;
Conference_Titel :
Consumer Electronics (GCCE), 2014 IEEE 3rd Global Conference on
Conference_Location :
Tokyo
DOI :
10.1109/GCCE.2014.7031263
