EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

On-Board Small-Scale Object Detection for Unmanned Aerial Vehicles (UAVs)

Zubair Saeed, Muhammad Haroon Yousaf, Rehan Ahmed, Sergio A. Velastin, Serestina Viriri

Drones · 06.05.2023 · 7 (5) · 310

DOI: 10.3390/drones7050310 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Drones, online erschienen am 06.05.2023. Verfasst von Zubair Saeed und 4 weiteren Autor:innen. Die Arbeit wurde bisher 43-mal zitiert (Crossref) und nennt 40 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Higher Education Commission of Pakistan funding.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Object detection is a critical task that becomes difficult when dealing with onboard detection using aerial images and computer vision technique. The main challenges with aerial images are small target sizes, low resolution, occlusion, attitude, and scale variations, which affect the performance of many object detectors. The accuracy of the detection and the efficiency of the inference are always trade-offs. We modified the architecture of CenterNet and used different CNN-based backbones of ResNet18, ResNet34, ResNet50, ResNet101, ResNet152, Res2Net50, Res2Net101, DLA-34, and hourglass14. A comparison of the modified CenterNet with nine CNN-based backbones is conducted and validated using three challenging datasets, i.e., VisDrone, Stanford Drone dataset (SSD), and AU-AIR. We also implemented well-known off-the-shelf object detectors, i.e., YoloV1 to YoloV7, SSD-MobileNet-V2, and Faster RCNN. The proposed approach and state-of-the-art object detectors are optimized and then implemented on cross-edge platforms, i.e., NVIDIA Jetson Xavier, NVIDIA Jetson Nano, and Neuro Compute Stick 2 (NCS2). A detailed comparison of performance between edge platforms is provided. Our modified CenterNet combination with hourglass as a backbone achieved 91.62%, 75.61%, and 34.82% mAP using the validation sets of AU-AIR, SSD, and VisDrone datasets, respectively. An FPS of 40.02 was achieved using the ResNet18 backbone. We also compared our approach with the latest cutting-edge research and found promising results for both discrete GPU and edge platforms.

Förderung

  • Higher Education Commission of Pakistan funding

Zitieren

APA

Saeed, Z., Yousaf, M. H., Ahmed, R., Velastin, S. A., & Viriri, S. (2023). On-Board Small-Scale Object Detection for Unmanned Aerial Vehicles (UAVs). Drones, 7(5), 310. https://doi.org/10.3390/drones7050310

Vancouver

Saeed Z, Yousaf MH, Ahmed R, Velastin SA, Viriri S. On-Board Small-Scale Object Detection for Unmanned Aerial Vehicles (UAVs). Drones. 2023;7(5):310. doi:10.3390/drones7050310

BibTeX
@article{Saeed2023,
  author = {Saeed, Zubair and Yousaf, Muhammad Haroon and Ahmed, Rehan and Velastin, Sergio A. and Viriri, Serestina},
  title = {On-Board Small-Scale Object Detection for Unmanned Aerial Vehicles (UAVs)},
  journal = {Drones},
  year = {2023},
  volume = {7},
  number = {5},
  pages = {310},
  doi = {10.3390/drones7050310}
}
Kostenlos starten →