Real-time Sign Detection and Recognition for Self-driving Mini Rovers based on Template Matching and Hierarchical Decision Structure

Quang Nhat Nguyen Le, Abir Bhattacharyya, Mamen Thomas Chembakasseril, Ronny Hartanto

2020

Abstract

Sign detection and recognition play vital roles in the field of self-driving vehicles. The aim of this paper is to introduce a real-time methodology that can be implemented on affordable single-board computers to classify varied traffic signs from the camera feed. The approach is to detect and recognise the colour and shape of the signs at first, then use the acquired information to access a hierarchy structure of the database in order to extract features of pre-existing templates. Finally, the template matching algorithm is applied to compare those features with potential Region Of Interest (ROI) based on a threshold value. We installed our system on a mini rover and tested it on a control urban traffic scenario. The measurements showed processing time ranging from 230ms to 800ms and 480ms to 1900ms on Jetson Nano and Raspberry Pi 3 Model B+ respectively.

Download


Paper Citation


in Harvard Style

Le Q., Bhattacharyya A., Chembakasseril M. and Hartanto R. (2020). Real-time Sign Detection and Recognition for Self-driving Mini Rovers based on Template Matching and Hierarchical Decision Structure. In Proceedings of the 12th International Conference on Agents and Artificial Intelligence - Volume 1: ICAART, ISBN 978-989-758-395-7, pages 208-215. DOI: 10.5220/0008969702080215


in Bibtex Style

@conference{icaart20,
author={Quang Le and Abir Bhattacharyya and Mamen Chembakasseril and Ronny Hartanto},
title={Real-time Sign Detection and Recognition for Self-driving Mini Rovers based on Template Matching and Hierarchical Decision Structure},
booktitle={Proceedings of the 12th International Conference on Agents and Artificial Intelligence - Volume 1: ICAART,},
year={2020},
pages={208-215},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0008969702080215},
isbn={978-989-758-395-7},
}


in EndNote Style

TY - CONF

JO - Proceedings of the 12th International Conference on Agents and Artificial Intelligence - Volume 1: ICAART,
TI - Real-time Sign Detection and Recognition for Self-driving Mini Rovers based on Template Matching and Hierarchical Decision Structure
SN - 978-989-758-395-7
AU - Le Q.
AU - Bhattacharyya A.
AU - Chembakasseril M.
AU - Hartanto R.
PY - 2020
SP - 208
EP - 215
DO - 10.5220/0008969702080215