NaSTSArLaT:Natural Scene Traffic Sign and Panel Guide Arabic-Latin Text

Citation Author(s):
Ridha
Zaghdoud
Khalil
Boukthir
Tarek M.
Hamdani
Adel M.
Alimi
Submitted by:
ridha zaghdoud
Last updated:
Wed, 10/05/2022 - 08:13
DOI:
10.21227/qhrx-1e76
Link to Paper:
License:
0
0 ratings - Please login to submit your rating.

Abstract 

The detection and recognition of road traffic signs and panel guide content has become challenging in recent years. Few studies have been made to solve these two issues at the same time, especially in the Arabic language. Additionally, the limited number of datasets for traffic signs and panel guide content makes the investigation more interesting. the Tunisian research groups in intelligent machines of the University of Sfax (REGIM laboratory of Sfax) will provide the NaSTSArLaT dataset free to researchers in traffic detection signs and traffic road scene text detection. The defined dataset contains  3000 collected images from google street views especially in Tunisia.

Instructions: 

Download the dataset and extract it.