Road Sign Recognition Based on Shape and Color Using SURF Method for Driver Assistance System
Year: 2018
Type: article
Abstract: The world is digitalizing day by day because of the contribution of science and technology.Intelligent Transport System (ITS) is a prominent innovation of this technology.It is an outstanding system which contributes not only to the betterment of human lives and environment but also to save time.Road sign recognition is one of the prominent sectors in the ITS.It provides real-time data on traffic for processing to take proper actions.This paper represents a color and shape based road sign detection model.As most of the road signs in Bangladesh are red; so, red channel of the image is extracted for detecting road sign accurately.A robust binarization technique is used to find the largest blob which indicates the region of interest (ROI).To extract ROI, the bounding box is used which is imposed on the RGB image.For decreasing computation time, the grayscale color space is used here.Then the features of the image are extracted and recognized by matching with the images in the database using Speed Up Robust Features (SURF) method.This method provides more accurate results in recognizing road signs.Various road sign images are used with a variety of conditions to test the proposed framework and results are presented to prove its effectiveness.
Language: English
Subfield: Automotive Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: bronze
OpenAlex ID: W2811187347