Improving Smart Cities Safety Using Sound Events Detection Based on Deep Neural Network Algorithms

In recent years, security in urban areas has gradually assumed a central position, focusing increasing attention on citizens, institutions and political forces. Security problems have a different nature—to name a few, we can think of the problems deriving from citizens’ mobility, then move on to mic...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Informatics (Basel) Jg. 7; H. 3; S. 23
Hauptverfasser: Ciaburro, Giuseppe, Iannace, Gino
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Basel MDPI AG 01.07.2020
Schlagworte:
Acoustics
Aircraft
Algorithms
Artificial intelligence
Artificial neural networks
audio events detection
Classification
convolutional neural networks
deep learning
Indoor environments
Internet of Things
Machine learning
Neural networks
pattern recognition
Public safety
Security
Sensors
Smart cities
Sound
sounds classification
Spectral signatures
Terrorism
Unmanned aerial vehicles
Urban areas
ISSN:2227-9709, 2227-9709
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:In recent years, security in urban areas has gradually assumed a central position, focusing increasing attention on citizens, institutions and political forces. Security problems have a different nature—to name a few, we can think of the problems deriving from citizens’ mobility, then move on to microcrime, and end up with the ever-present risk of terrorism. Equipping a smart city with an infrastructure of sensors capable of alerting security managers about a possible risk becomes crucial for the safety of citizens. The use of unmanned aerial vehicles (UAVs) to manage citizens’ needs is now widespread, to highlight the possible risks to public safety. These risks were then increased using these devices to carry out terrorist attacks in various places around the world. Detecting the presence of drones is not a simple procedure given the small size and the presence of only rotating parts. This study presents the results of studies carried out on the detection of the presence of UAVs in outdoor/indoor urban sound environments. For the detection of UAVs, sensors capable of measuring the sound emitted by UAVs and algorithms based on deep neural networks capable of identifying their spectral signature that were used. The results obtained suggest the adoption of this methodology for improving the safety of smart cities.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2227-9709
2227-9709
DOI:10.3390/informatics7030023