Geometric Flood Depth Estimation: Fusing Transformer-Based Segmentation with Digital Elevation Models

TL;DR

This paper introduces a geometric method combining transformer-based segmentation with digital elevation models to estimate flood depth from aerial images, enhancing post-disaster situational awareness.

Contribution

It presents a novel pipeline that fuses high-accuracy flood masks with elevation data to derive 3D flood depth without hydrodynamic simulations.

Findings

Accurate flood masks generated by Mask2Former improve depth estimation.

Fused data enables calculation of water surface elevation and per-pixel depth.

Method evaluated successfully on FloodNet and CRASAR-U-DROIDS datasets.

Abstract

Post-disaster situational awareness relies heavily on understanding both the extent and the volume of floodwaters. While 2D semantic segmentation provides accurate flood masking, it lacks the vertical dimension required to assess navigability and structural risk. This paper presents a geometric "Water Surface Elevation" approach for estimating flood depth from monocular aerial imagery. Our pipeline utilizes Mask2Former, a state-of-the-art transformer-based segmentation model, to generate precise 2D flood masks. These masks are fused with Digital Elevation Models (DEMs) to identify the water-land boundary, calculate a global water surface elevation ($Z_{water}$), and compute per-pixel depth based on the principle of local hydrostatic equilibrium. We evaluate this workflow using the FloodNet and CRASAR-U-DROIDS datasets, demonstrating how high-performance segmentation can be leveraged to extract 3D volumetric data from 2D imagery without the latency of hydrodynamic simulations.