Abstract
This paper proposes a computer vision based method for traffic sign detection
recognition and tracking based on the color and shape of the road sign and its geometric
attributes. The study of traffic sign detection has been of great interests and often addressed
by a three-stage procedure involving detection, tracking and classification. Road safety is an
issue of national concern and its impacts is on the economy, public health and the
general welfare of the people. We report on-going efforts to develop an intelligent agent
for detecting and tracking traffic signs for vision based Driver Assistance System (DAS).
Initially paper describes general framework for traffic sign detection and their important
subsystems. Generally, computer vision techniques consist of three important stages, color
segmentation, sign detection and classification. We explore the important sub-systems of
each of these stages, and identify their advantages and disadvantages. We propose a system
based on the framework to detect circular and triangular road traffic signs making use of
artificial intelligence techniques such as heuristics functions for detecting shapes. System is
implemented in real time environment and tested on national highway and aims to help
vehicle driver to have safer and enjoyable driving thereby concentrating on his actual
workload.
Users
Please
log in to take part in the discussion (add own reviews or comments).