Geographical Hidden Markov Tree for Flood Extent Mapping (With Proof Appendix)
This addresses flood mapping for disaster management and water forecasting, but it appears incremental as it builds on existing hidden Markov models with spatial adaptations.
The authors tackled flood extent mapping by proposing a geographical hidden Markov tree model that generalizes hidden Markov models to two-dimensional maps with anisotropic spatial dependency, showing it outperforms multiple baselines on synthetic and real-world datasets with scalable algorithms.
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.