Drone Forensics and Machine Learning: Sustaining the Investigation Process

Drones have been increasingly adopted to address several critical challenges faced by humanity to provide support and convenience . The technological advances in the broader domains of artificial intelligence and the Internet of Things (IoT) as well as the affordability of off-the-shelf devices, hav...

Full description

Saved in:
Bibliographic Details
Published inSustainability Vol. 14; no. 8; p. 4861
Main Authors Baig, Zubair, Khan, Majid Ali, Mohammad, Nazeeruddin, Brahim, Ghassen Ben
Format Journal Article
LanguageEnglish
Published Basel MDPI AG 01.04.2022
Subjects
Access control
Artificial intelligence
Computer forensics
Crime
Criminal investigations
Data collection
Drones
Evidence
Forensic science
Forensic sciences
Global positioning systems
GPS
Image retrieval
Internet of Things
Law enforcement
Learning algorithms
Machine learning
Meteorological data
Radio communications
Smartphones
Software
Surveillance
Unmanned aerial vehicles
Online AccessGet full text

Cover

More Information
Summary:Drones have been increasingly adopted to address several critical challenges faced by humanity to provide support and convenience . The technological advances in the broader domains of artificial intelligence and the Internet of Things (IoT) as well as the affordability of off-the-shelf devices, have facilitated modern-day drone use. Drones are readily available for deployment in hard to access locations for delivery of critical medical supplies, for surveillance, for weather data collection and for home delivery of purchased goods. Whilst drones are increasingly beneficial to civilians, they have also been used to carry out crimes. We present a survey of artificial intelligence techniques that exist in the literature in the context of processing drone data to reveal criminal activity. Our contribution also comprises the proposal of a novel model to adopt the concepts of machine learning for classification of drone data as part of a digital forensic investigation. Our main conclusions include that properly trained machine-learning models hold promise to enable an accurate assessment of drone data obtained from drones confiscated from a crime scene. Our research work opens the door for academics and industry practitioners to adopt machine learning to enable the use of drone data in forensic investigations.
ISSN:2071-1050
2071-1050
DOI:10.3390/su14084861