Hybrid hidden Markov model for marine environment monitoring

Type Article
Date 2015-01
Language English
Author(s) Rousseeuw Kevin1, 2, Poison Caillault Emilie1, Lefebvre AlainORCID2, Hamad Denis1
Affiliation(s) 1 : ULCO/LISIC, BP 719, FR-62228 Calais, France
2 : IFREMER, Centre Manche Mer du Nord, BP 699, FR-62321 Boulogne-sur-Mer, France
Source Ieee Journal Of Selected Topics In Applied Earth Observations And Remote Sensing (1939-1404) (Institute of Electrical and Electronics Engineers (IEEE)), 2015-01 , Vol. 8 , N. 1 , P. 204-213
DOI 10.1109/JSTARS.2014.2341219
WOS© Times Cited 6
Keyword(s) Hybrid Hidden Markov Model, marine water monitoring, Phytoplankton blooms, spectral clustering
Abstract Phytoplankton is an important indicator of water quality assessment. To understand phytoplankton dynamics, many fixed buoys and ferry boxes were implemented, resulting in the generation of substantial data signals. Collected data are used as inputs of an effective monitoring system. The system, based on unsupervised hidden Markov model (HMM), is designed not only to detect phytoplancton blooms but also to understand their dynamics. HMM parameters are usually estimated by an iterative expectation-maximization (EM) approach. We propose to estimate HMM parameters by using spectral clustering algorithm. The monitoring system is assessed based on database signals from MAREL-Carnot station, Boulogne-sur-Mer, France. Experimental results show that the proposed system is efficient to detect environmental states such as phytoplankton productive and nonproductive periods without a priori knowledge. Furthermore, discovered states are consistent with biological interpretation.
Full Text
File Pages Size Access
Author's final draft 13 577 KB Open access
Top of the page