EUVIMEDEUVIMED entdecken
Crossref

Design and Practice of Efficient Acceleration System for YOLOv8 Object Detection Algorithm Based on FPGA

Qicheng Wen

Journal of Intelligence and Knowledge Engineering · 2026-06 · 4 (2) · 183-189

DOI: 10.62517/jike.202604225 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Intelligence and Knowledge Engineering, 2026-06. Verfasst von Qicheng Wen. Die Arbeit nennt 21 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This paper addresses the performance bottleneck of the YOLOv8 object detection algorithm in real-time applications by proposing an efficient FPGA-based acceleration system design. First, the system reviews the current state of object detection algorithms and FPGA technology, analyzing the shortcomings of existing FPGA acceleration solutions. Second, at the system design level, it focuses on optimization strategies for the YOLOv8 algorithm, including network structure simplification and computational flow reconstruction, and completes the selection and configuration design of the FPGA hardware platform. In the implementation phase, the paper elaborates on the algorithm’s migration process to FPGA and the hardware circuit implementation, including PCB design. Experimental results demonstrate that the designed acceleration system significantly improves processing speed while maintaining detection accuracy, validating the effectiveness of FPGA hardware acceleration. Finally, the paper discusses key factors influencing system performance optimization and identifies future improvement directions, such as algorithm-hardware co-design and dynamic reconfiguration. This study provides an efficient hardware acceleration solution for real-time object detection applications, offering significant theoretical value and practical implications.

Zitieren

APA

Wen, Q. (2026). Design and Practice of Efficient Acceleration System for YOLOv8 Object Detection Algorithm Based on FPGA. Journal of Intelligence and Knowledge Engineering, 4(2), 183-189. https://doi.org/10.62517/jike.202604225

Vancouver

Wen Q. Design and Practice of Efficient Acceleration System for YOLOv8 Object Detection Algorithm Based on FPGA. Journal of Intelligence and Knowledge Engineering. 2026;4(2):183-189. doi:10.62517/jike.202604225

BibTeX
@article{Wen2026,
  author = {Wen, Qicheng},
  title = {Design and Practice of Efficient Acceleration System for YOLOv8 Object Detection Algorithm Based on FPGA},
  journal = {Journal of Intelligence and Knowledge Engineering},
  year = {2026},
  volume = {4},
  number = {2},
  pages = {183--189},
  doi = {10.62517/jike.202604225}
}
Kostenlos starten →