[M-GI31] Earth and planetary informatics and data utilization

[MGI31-04] Towards Active Flood Risk Assessment and Prediction through Automated River Level Monitoring

*Francis Tiausas¹, Ken T. Murata¹, Kazuki Kobayashi¹ (1.Shinshuu University)

Keywords:Computer Vision, River Monitoring, PTZ Cameras

Due to global warming, the occurrence of unexpected weather events becomes increasingly common. This is a big problem for areas like riverside towns and villages which have no previous experience of severe flooding as they are more likely to be inadequately prepared to handle such scenarios. Therefore, digital systems for the early detection and continuous monitoring of the current state of the river can give crucial information to local governments and the general public. This paper presents the progress of a River Water Level Monitoring System using hourly videos captured by a Pan-Tilt-Zoom (PTZ) camera of several points along the Chikuma River. A novel combination of image processing techniques are then used to extract the current shape and edge of the river from the given video which can then be used to determine the current river water level. The end goal is to be able to produce active flood risk assessment and prediction models based on the output of the system.