Real-Time Video Road Sign Detection and Tracking Using Image Processing and Autonomous Car
F.A. Azis1, P.S.G. Ponaseran2, Zamani Md Sani3, M.S.M. Aras4, M. Nur Othman5
1Fadilah Abdul Azis, Faculty of Electrical Engineering, Universiti Teknikal Malaysia Melaka, Melaka, Malaysia.
2Zamani Md Sani, Faculty of Electrical Engineering, Universiti Teknikal Malaysia Melaka, Melaka, Malaysia.
3Mohd Shahrieel Mohd Aras, Faculty of Electrical Engineering, Universiti Teknikal Malaysia Melaka, Melaka, Malaysia.
4M. Nur Othman, Faculty of Mechanical and Manucfacturing Technology Engineering, Universiti Teknikal Malaysia Melaka, Melaka, Malaysia.
5P.S.G. Ponaseran, Faculty of Electrical Engineering, Universiti Teknikal Malaysia Melaka, Melaka, Malaysia.
Manuscript received on 09 December 2019 | Revised Manuscript received on 23 December 2019 | Manuscript Published on 31 December 2019 | PP: 495-500 | Volume-8 Issue-12S2 October 2019 | Retrieval Number: L109410812S219/2019©BEIESP | DOI: 10.35940/ijitee.L1094.10812S219
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open-access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Detection and monitoring of real-time road signs are becoming today’s study in the autonomous car industry. The number of car users in Malaysia risen every year as well as the rate of car crashes. Different types, shapes, and colour of road signs lead the driver to neglect them, and this attitude contributing to a high rate of accidents. The purpose of this paper is to implement image processing using the real-time video Road Sign Detection and Tracking (RSDT) with an autonomous car. The detection of road signs is carried out by using Video and Image Processing technique control in Python by applying deep learning process to detect an object in a video’s motion. The extracted features from the video frame will continue to template matching on recognition processes which are based on the database. The experiment for the fixed distance shows an accuracy of 99.9943% while the experiment with the various distance showed the inversely proportional relation between distances and accuracies. This system was also able to detect and recognize five types of road signs using a convolutional neural network. Lastly, the experimental results proved the system capability to detect and recognize the road sign accurately.
Keywords: Convolutional Neural Network, Image Processing, Real-Time, Road Signs.
Scope of the Article: Signal and Image Processing