Afficher la notice abrégée
dc.contributor.author |
Keddous, Akila;
Lagha, Mohand (promoteur);
Choutri, Kheireddine (promoteur) |
|
dc.date.accessioned |
2021-12-06T10:21:12Z |
|
dc.date.available |
2021-12-06T10:21:12Z |
|
dc.date.issued |
2021 |
|
dc.identifier.uri |
http://di.univ-blida.dz:8080/jspui/handle/123456789/13403 |
|
dc.description |
Mémoire de Master option Avionique.-Numéro de Thèse 054/2021 |
fr_FR |
dc.description.abstract |
Forest fires are very dangerous. Once they become widespread, it is very difficult to extinguish.
In this work, an Unmanned aerial vehicle (UAV) image-based Real-time Forest fire detectionapproach is proposed. Where we took advantage of recent development in computer vision
systems and the rapid maneuverability of Unmanned Aerial Vehicles to improve the performance of the Real time detection, we designed and implemented a YOLOv2Convolutional Neural Network Model in MATLAB to train on an aerial dataset, Experimentalresults show that our proposed system has high detection performance, and its detection speedreaches 58 Frame Per Second with a mean average precision of 0.87, thereby satisfying therequirements of real-time detection (Speed and Accuracy). |
fr_FR |
dc.language.iso |
en |
fr_FR |
dc.publisher |
Université Blida 01 |
fr_FR |
dc.subject |
Real-time fire detection;
Deep learning;
Convolutional neural network;
Computer vision;
Unmanned aerial vehicles;
Fire Datasets;
YOLOv2 |
fr_FR |
dc.title |
UAV aerial image-based forest fire detection using deep learning |
fr_FR |
dc.type |
Thesis |
fr_FR |
Fichier(s) constituant ce document
Ce document figure dans la(les) collection(s) suivante(s)
Afficher la notice abrégée