Urban flood risk is intensifying due to climate change and high-intensity rainfall, especially in cities with limited drainage capacity.

• Accurate and time-resolved flood-depth information is critical for early warning, emergency response, and disaster risk reduction.

• This study proposes a near real-time flood monitoring framework that integrates image-based flood detection using YOLOv8-seg with a two-dimensional hydrodynamic model (PCSWMM) to improve inundation mapping accuracy and warning reliability.

This poster was presented during the 11th APN Early Career Professional Poster and Networking Session, held alongside the APN 27th Joint Intergovernmental and Scientific Planning Group Meetings on 9 July 2026 in Bangkok, Thailand.