Please use this identifier to cite or link to this item:
Title: Vehicular dataset for road assessment conditions
Author: Antunes, Mário
Gomes, Diogo
Barraca, João Paulo
Aguiar, Rui L.
Keywords: IoT
Machine Learning
Issue Date: 2017
Publisher: IEEE
Abstract: The Internet of Things (IoT) is a very promising concept that by connecting numerous devices to the internet and extracting large sums of information (BigData) can enable the realisation of various futuristic scenarios. In order to develop and assess future applications and services, it is necessary the availability of datasets that can be used to train, test and cross validate. Project SCoT (Smart Cloud of Things) has developed an M2M platform capable of collecting information from heterogeneous devices and collide that information in a large data repository. During its pilot phase, the project made the assessment of the road conditions in the region of Aveiro, Portugal. In this work we make the dataset used on the previous mentioned pilot publicly available. With this dataset our road assessment algorithm reached 80$\backslash$$\backslash${\%} accuracy in the task of pothole detection, other scenarios (that take into account vehicular speed, position and acceleration) can also be explored. The dataset was not pre-processed in anyway, the only transformation was made to protect the identity of the volunteers.
Peer review: yes
DOI: 10.1109/ISC2.2017.8090867
ISBN: 978-1-5386-2524-8
Appears in Collections:DETI - Comunicações
IT - Comunicações

Files in This Item:
File Description SizeFormat 
paper.pdf4.82 MBAdobe PDFView/Open

Formato BibTex MendeleyEndnote Degois 

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.