Joint classification of multiresolution and multisensor data using a multiscale Markov mesh model

Alessandro Montaldo (1) , Luca Fronda (1) , Ihsen Hedhli (2, 3) , Gabriele Moser (1) , Josiane Zerubia (3, 4) , Sebastiano B. Serpico (1)

1 UniGe - Università degli studi di Genova = University of Genoa
2 ULaval - Université Laval [Québec]
3 CRISAM - Inria Sophia Antipolis - Méditerranée
4 AYIN - Models of spatio-temporal structure for high-resolution image processing

In this paper, the problem of the classification of multireso-lution and multisensor remotely sensed data is addressed by proposing a multiscale Markov mesh model. Multiresolution and multisensor fusion are jointly achieved through an explicitly hierarchical probabilistic graphical classifier, which uses a quadtree structure to model the interactions across different spatial resolutions, and a symmetric Markov mesh random field to deal with contextual information at each scale and favor applicability to very high resolution imagery. Differently from previous hierarchical Markovian approaches, here, data collected by distinct sensors are fused through either the graph topology itself (across its layers) or decision tree ensemble methods (within each layer). The proposed model allows taking benefit of strong analytical properties, most remarkably causality, which make it possible to apply time-efficient non-iterative inference algorithms.

Mots clés: hierarchical MRF; symmetric Markov mesh; Multiresolution and multisensor fusion; Quad tree

Domaines: Informatique [cs]; Apprentissage [cs.LG]; Mathématiques [math]

Alessandro Montaldo, Luca Fronda, Ihsen Hedhli, Gabriele Moser, Josiane Zerubia, et al.. Joint classification of multiresolution and multisensor data using a multiscale Markov mesh model. IGARSS 2019 - IEEE International Geoscience and Remote Sensing Symposium, Jul 2019, Yokohama, Japan. ⟨hal-02157082⟩