EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A novel deep learning-based object detection along semantic segmentation on aerial imagery

Maha Abdelhaq, Aysha Naseer, Raed Alsaqour, Ahmad Jalal, Muhammad Asif Jamal, Aamir Nadeem, Jinsul Kim

PeerJ Computer Science · 24.03.2026 · 12 · e3648

DOI: 10.7717/peerj-cs.3648 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PeerJ Computer Science, online erschienen am 24.03.2026. Verfasst von Maha Abdelhaq und 6 weiteren Autor:innen. Die Arbeit nennt 44 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Innovative Human Resource Development for Local Intellectualization Program, The Institute of Information & Communications Technology Planning & Evaluation, Korea Government.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Deep learning has significantly advanced object detection and scene understanding, yet analyzing uncrewed aerial vehicle (UAV)-based hyperspectral imagery remains challenging due to its high spectral complexity and cluttered scenes. Most existing approaches address either detection or segmentation separately, or rely on limited data modalities, which constrains their effectiveness in complex aerial scenarios. To address this gap, we propose AeroResNet-Vision (ARV), a novel multimodal fusion framework for UAV imagery. ARV integrates state-of-the-art techniques into a unified pipeline: a ResNet-based DeepLabv3++ module for multi-scale semantic segmentation, classical Binary Robust Invariant Scalable Keypoints (BRISK), Maximally stable Extrema Regions (MSER) detectors for keypoint and region feature extraction, (You Only Look Once, version 8) YOLOv8 for efficient object detection, and a Vision Transformer (ViT) for context-aware classification. Additionally, image normalization and edge detection preprocessing are applied to enhance image quality and emphasize structural features. By fusing handcrafted and deep features from these components, ARV effectively handles varying object scales and complex backgrounds in aerial data. Experimental results on three benchmark datasets (ISPRS Potsdam, VEDAI, UAVid) demonstrate state-of-the-art accuracy—for instance, ARV achieved 97.20%, 98.50%, and 97.60% accuracy on these datasets, respectively. These findings validate the framework’s superior performance in UAV image analysis. In conclusion, the proposed multimodal approach provides a robust solution for aerial object recognition, and we recommend exploring lightweight model variants and self-supervised learning to further enhance its deployment potential.

Förderung

  • Innovative Human Resource Development for Local Intellectualization Program
  • The Institute of Information & Communications Technology Planning & Evaluation
  • Korea Government
  • Princess Nourah bint Abdulrahman University Researchers, Riyadh, Saudi Arabia

Zitieren

APA

Abdelhaq, M., Naseer, A., Alsaqour, R., Jalal, A., Jamal, M. A., Nadeem, A., & Kim, J. (2026). A novel deep learning-based object detection along semantic segmentation on aerial imagery. PeerJ Computer Science, 12, e3648. https://doi.org/10.7717/peerj-cs.3648

Vancouver

Abdelhaq M, Naseer A, Alsaqour R, Jalal A, Jamal MA, Nadeem A, et al. A novel deep learning-based object detection along semantic segmentation on aerial imagery. PeerJ Computer Science. 2026;12:e3648. doi:10.7717/peerj-cs.3648

BibTeX
@article{Abdelhaq2026,
  author = {Abdelhaq, Maha and Naseer, Aysha and Alsaqour, Raed and Jalal, Ahmad and Jamal, Muhammad Asif and Nadeem, Aamir and Kim, Jinsul},
  title = {A novel deep learning-based object detection along semantic segmentation on aerial imagery},
  journal = {PeerJ Computer Science},
  year = {2026},
  volume = {12},
  pages = {e3648},
  doi = {10.7717/peerj-cs.3648}
}
Kostenlos starten →