EUVIMEDEUVIMED entdecken
Open Access · CC BY-SA 4.0

Vision-Based Water Level Tracking: Integrating YOLOv8 and ByteTrack for Early Flood Warning

Soheer Ahmed

Journal of Applied Technology Research and Innovation · 2026

DOI: 10.68184/rdppvh33 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Applied Technology Research and Innovation, 2026. Verfasst von Soheer Ahmed. Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Floods rank among the most devastating natural disasters worldwide, demanding robust monitoring systems capable of delivering timely, low-latency warnings. Conventional approaches relying on manual observation and physical sensor networks suffer from high deployment costs, limited spatial coverage, and response times too slow to support effective early warning. This work describes a vision-based flood monitoring system that uses YOLOv8, trained on two custom-annotated video datasets, Python, OpenCV, and the ByteTrack tracking algorithm to identify rising water levels in real time. When water levels exceed a predetermined threshold, the system automatically sends out on-screen alerts and email notifications using SMTP. In a comparative test, YOLOv8's anchor-free design surpasses YOLOv5 in real-time frame processing, particularly for dynamic water bodies. The proposed system offers a scalable, cost-effective solution suitable for deployment at rivers, dams, and glacial melt zones, with the potential to significantly reduce flood-related fatalities in under-resourced regions.

Zitieren

APA

Ahmed, S. (2026). Vision-Based Water Level Tracking: Integrating YOLOv8 and ByteTrack for Early Flood Warning. Journal of Applied Technology Research and Innovation. https://doi.org/10.68184/rdppvh33

Vancouver

Ahmed S. Vision-Based Water Level Tracking: Integrating YOLOv8 and ByteTrack for Early Flood Warning. Journal of Applied Technology Research and Innovation. 2026. doi:10.68184/rdppvh33

BibTeX
@article{Ahmed2026,
  author = {Ahmed, Soheer},
  title = {Vision-Based Water Level Tracking: Integrating YOLOv8 and ByteTrack for Early Flood Warning},
  journal = {Journal of Applied Technology Research and Innovation},
  year = {2026},
  doi = {10.68184/rdppvh33}
}
Kostenlos starten →