Sign in Subscribe
Algorithms

Machine Learning Algorithm Combining AI and Edge Detection Advances Autonomous Flood Observation Systems

Strategic Insights

— 2 min read

Researchers from the Virginia Institute of Marine Science have successfully integrated passive remote sensors with cutting-edge machine learning algorithms and edge detection to accelerate the development of autonomous flood observation systems. This innovative approach leverages passive remote sensing to capture free surface water levels from live oblique pictometry in tidal water bodies. The study, funded by the USGS Next Generation Water Observing System (NGWOS) Research and Development program, deployed live-streaming web cameras to collect images at six-minute intervals over a three-month period. Statistical analysis revealed that these sensors provided continuous surface water level measurements with a root mean square error (RMSE) deviation of 0.55 meters. The research has been peer-reviewed and published in the journal Applied Ocean Research.

Key Takeaways:

  • The researchers developed an advanced machine learning algorithm combining AI and edge detection to enhance autonomous flood observation systems.
  • The algorithm leveraged passive remote sensing to capture free surface water levels from live oblique pictometry in tidal water bodies.
  • The study deployed live-streaming web cameras to collect images at six-minute intervals over a three-month period.
  • Statistical analysis demonstrated that the sensors provided continuous surface water level measurements with a RMSE deviation of 0.55 meters.
  • The research has been funded by the USGS Next Generation Water Observing System (NGWOS) Research and Development program.
  • The study has been peer-reviewed and published in the journal Applied Ocean Research.
  • The research team included J. Derek Loftis, Hunter Harman, Sridhar Katragadda, and Russell Lotspeich from the Virginia Institute of Marine Science.
  • The study aims to improve the accuracy and efficiency of flood observation systems, which is crucial for predicting and mitigating the impacts of flooding.

Statistics:

  • 6-minute intervals: the frequency at which live-streaming web cameras collected images during the 3-month study period.
  • 3 months: the duration of the study period during which live-streaming web cameras collected images.
  • 0.55 meters: the RMSE deviation of surface water level measurements provided by the sensors.
  • USGS Next Generation Water Observing System (NGWOS) Research and Development program: the funding agency for the research study.
  • 1375 Greate Rd: the address of the Virginia Institute of Marine Science, where the research team is based.

Sources:

  • Inundation Monitoring Using a Machine Learning Algorithm Combining Ai and Edge Detection. Applied Ocean Research, 2025;160. (Elsevier Sci Ltd)
  • Findings from Virginia Institute of Marine Science Reveals New Findings on Machine Learning (Inundation Monitoring Using a Machine Learning Algorithm Combining Ai and Edge Detection). Robotics & Machine Learning. July 7, 2025; p 181. (NewsRx LLC)

Read more