Real-Time Video Road Sign Detection And Tracking Using Image Processing And Autonomous Car

Ponaseran, P.S.Giritharan (2018) Real-Time Video Road Sign Detection And Tracking Using Image Processing And Autonomous Car. Project Report. UTeM, Melaka, Malaysia. (Submitted)

Text (24 Pages) Real-Time Video Road Sign Detection And Tracking Using Image Processing And Autonomous Car.pdf - Submitted Version Download (1MB)

Abstract

Road Sign Detection and Tracking (RSDT) system becomes one of today’s research in autonomous car industries. A safe driving becomes a priority for all drivers and passengers. The rate of cars user increased by each year and it’s also increasing the rate of accidents in Malaysia. Neglecting road signs contributing to high rate of accidents. Drivers get confused because there are many types of road signs with different types of shapes and colours. Besides, drivers missed the road signs because of not giving full attention while driving. The research aiming to improve the safety in driving among the driver in road. The system is design to classify 5 types of road signs. Real-time video RSDT is a system that will detect and classify a road sign on a video’s motion by using image processing. RSDT will help the drivers to recognise the road signs while driving image alert description. The detection of road signs on the motion video is done by using Video and Image Processing technique control in Python by applying deep learning process to detect an object in video’s motion. The features such as road sign that is successfully extracted from the video frame will be proceed to template matching on recognition process based on the template in the database. The experiment for the fixed distance shows an accuracy of 99.9943% while the experiment with various distance the accuracy decreased as the distance increasing. For the distance of 4 metre from the road sign to camera the accuracy is 99.9842%, accuracy for 5 metre is 99.9780% and for 6 metre is 99.97114%. For the light intensity on morning the accuracy is 99.9934% and on afternoon the accuracy is 99.9966%. The result showing, the system able to detect and recognise the road sign accurately and reliable. In a nutshell, RSDT will be a smart system in future for drivers to use while driving on road.

Actions (login required)

View Item