# Geographical Hidden Markov Tree for Flood Extent Mapping (With Proof Appendix)

arXiv: https://arxiv.org/abs/1805.09757 · HTML: https://mlsanity.com/paper/1805.09757/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 24 May 2018.

## Built on by (most cited first, 4 of 4)

- [FloodLense: A Framework for ChatGPT-based Real-time Flood Detection](https://mlsanity.com/paper/2401.15501/) (2024)
- [Spatial Knowledge-Infused Hierarchical Learning: An Application in Flood Mapping on Earth Imagery](https://mlsanity.com/paper/2312.07767/) (2023)
- [EvaNet: Elevation-Guided Flood Extent Mapping on Earth Imagery (Extended Version)](https://mlsanity.com/paper/2404.17917/) (2024)
- [Enabling Fast and Accurate Crowdsourced Annotation for Elevation-Aware Flood Extent Mapping](https://mlsanity.com/paper/2408.05350/) (2024)

## Abstract (arXiv)

Flood extent mapping plays a crucial role in disaster management and national water forecasting. Unfortunately, traditional classification methods are often hampered by the existence of noise, obstacles and heterogeneity in spectral features as well as implicit anisotropic spatial dependency across class labels. In this paper, we propose geographical hidden Markov tree, a probabilistic graphical model that generalizes the common hidden Markov model from a one dimensional sequence to a two dimensional map. Anisotropic spatial dependency is incorporated in the hidden class layer with a reverse tree structure. We also investigate computational algorithms for reverse tree construction, model parameter learning and class inference. Extensive evaluations on both synthetic and real world datasets show that proposed model outperforms multiple baselines in flood mapping, and our algorithms are scalable on large data sizes.