Markov Models and Extensions for Land Cover Mapping in Aerial Imagery

Mohamed El Yazid Boudaren (1) , Abdel Belaïd (2)

1 USTHB - Université des Sciences et de la Technologie Houari Boumediene = University of Sciences and Technology Houari Boumediene [Alger]
2 READ - READ

Markov models are well-established stochastic models for image analysis and processing since they allow one to take into account the contextual relationships between image pixels. In this paper, we attempt to methodically review the use of Markov models and their extensions for Land Cover mapping problem in aerial imagery according to available literature and previous research works. A new Markov model combining Markov random fields and hidden Markov models and inspired from the NSHP-HMM model, initially introduced for Handwritten Words Recognition is defined. New learning and labeling procedures are derived.

Mots clés: Aerial Images; Land Cover Mapping; Markov models; Markov Random Fields MRF; Hidden Markov Models HMM

Domaines: Traitement du signal et de l'image [eess.SP]; Traitement du signal et de l'image [eess.SP]

Mohamed El Yazid Boudaren, Abdel Belaïd. Markov Models and Extensions for Land Cover Mapping in Aerial Imagery. International Conference of Signal and Image Engineering - ICSIE 2009, Jul 2009, London, United Kingdom. ⟨inria-00346632⟩