A Comprehensive Study on Deep Learning Architectures for Robust Object Identification in UAV-based Thermal Imaging

DOI

10.2991/978-94-6239-664-7_44How to use a DOI?

Keywords

Computer Vision; Deep Learning; Object Detection; YOLO; UAV thermal Image

Abstract

In recent years, various industries have been experiencing a notable increase in the use of machine learning for object recognition, resulting in the development of different methods and technologies. Thermal detection technology is one such aspect that uses thermal images captured from objects to determine them based on their heat signatures. This has been particularly useful in situations like surveillance, search and rescue missions, and nighttime operations where conventional imaging techniques are limited by poor visibility or darkness. The advent of unmanned aerial vehicles (UAVs) has even widened the scope of practical application of thermal imaging. In situations such as monitoring wildfires or carrying out search and rescue operations, UAVs fitted with thermographic cameras can cover large areas within a short time. The combination of UAVs and thermal detection technology has greatly improved thermal data acquisition expanding its reach into fields such as disaster relief, animal tracking among others including industry inspections. In this study, we used the HIT-UAV dataset which has much popularity in object identification research to investigate the efficiency of the latest YOLOv8 model more accurately the YOLOv8m and YOLOv8s by comparing these models to its previous version YOLOv5 models. Through our implementation of YOLOv8m on the HIT-UAV dataset we were able to achieve precision of 87.6% and mean average score of 81.9% and with YOLOv8s we were able to achieve precision of 89.1% and mean average score of 82.3%.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Sadman Sadik Khan
AU  - Mahtab Chowdhury
AU  - Md Shahriar Mannan Prottoy
AU  - Kazi Zakiul Haque
AU  - Al Momit
AU  - Tasnuva Arfin Janisa
PY  - 2026
DA  - 2026/06/08
TI  - A Comprehensive Study on Deep Learning Architectures for Robust Object Identification in UAV-based Thermal Imaging
BT  - Proceedings of the International Conference on Intelligent Data Analysis and Applications (IDAA 2025)
PB  - Atlantis Press
SP  - 636
EP  - 647
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6239-664-7_44
DO  - 10.2991/978-94-6239-664-7_44
ID  - Khan2026
ER  -