EUVIMEDEUVIMED entdecken
Crossref

Advancing UAV Multi-object Tracking with YOLOv8, Nano Instance Segmentation and Dueling Double DQN

R. Kiruthiga, B. Nithya, S. Martin Prabhu

SN Computer Science · 11.08.2026 · 7 (6)

DOI: 10.1007/s42979-026-05266-0 ↗

Kurz gefasst

Fachartikel in der Zeitschrift SN Computer Science, online erschienen am 11.08.2026. Verfasst von R. Kiruthiga und 2 weiteren Autor:innen. Die Arbeit nennt 30 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Unmanned Aerial Vehicles (UAVs) have become essential for navigating complex terrains, accessing remote or hazardous locations, and capturing high-resolution imagery. This paper presents an advanced approach to object detection and tracking within the realm of computer vision for UAVs. Conventional deep learning models, such as RCNN, Fast RCNN, and YOLO, face challenges in handling occluded, blurred, or densely clustered objects and often struggle with simultaneous multi-object tracking. To overcome these limitations, the proposed framework integrates YOLOv8x (You Only Look Once), Nano Instance Segmentation (NIS), and Dueling Double Deep Q Network (DDDQN). YOLOv8x outperforms its predecessors, achieving an impressive Average Precision (AP) of 53.9% on the MSCOCO dataset’s challenging test set. The DDDQN algorithm further enhances tracking efficiency by independently estimating state values and state-dependent action advantages. The synergy between YOLOv8x and DDDQN enables robust handling of obstacles, varying object sizes, and unpredictable movements. The proposed framework is evaluated on the UAVDT and VisDrone datasets and compared against nine state-of-the-art methods. Results indicate significant improvements in object tracking performance, particularly in densely populated environments, positioning this framework as a promising solution for real-world applications requiring precise and resilient object detection.

Zitieren

APA

Kiruthiga, R., Nithya, B., & Prabhu, S. M. (2026). Advancing UAV Multi-object Tracking with YOLOv8, Nano Instance Segmentation and Dueling Double DQN. SN Computer Science, 7(6). https://doi.org/10.1007/s42979-026-05266-0

Vancouver

Kiruthiga R, Nithya B, Prabhu SM. Advancing UAV Multi-object Tracking with YOLOv8, Nano Instance Segmentation and Dueling Double DQN. SN Computer Science. 2026;7(6). doi:10.1007/s42979-026-05266-0

BibTeX
@article{Kiruthiga2026,
  author = {Kiruthiga, R. and Nithya, B. and Prabhu, S. Martin},
  title = {Advancing UAV Multi-object Tracking with YOLOv8, Nano Instance Segmentation and Dueling Double DQN},
  journal = {SN Computer Science},
  year = {2026},
  volume = {7},
  number = {6},
  doi = {10.1007/s42979-026-05266-0}
}
Kostenlos starten →