A Hierarchical Spatio-Temporal Markov Model for Improved Flood Mapping Using Multi-Temporal X-Band SAR Data

Remote Sensing Remote Sensing 2072-4292 RSBSBZ 9 2010 2 9 A Hierarchical Spatio-Temporal Markov Model for Improved Flood Mapping Using Multi-Temporal X-Band SAR Data Sandro Martinis German Aerospace Center (DLR), German Remote Sensing Data Center (DFD), Oberpfaffenhofen, 82234 Wessling, Germany André Twele German Aerospace Center (DLR), German Remote Sensing Data Center (DFD), Oberpfaffenhofen, 82234 Wessling, Germany

In this contribution, a hybrid multi-contextual Markov model for unsupervised near real-time flood detection in multi-temporal X-band synthetic aperture radar (SAR) data is presented. It incorporates scale-dependent, as well as spatio-temporal contextual information, into the classification scheme, by combining hierarchical marginal posterior mode (HMPM) estimation on directed graphs with noncausal Markov image modeling related to planar Markov random fields (MRFs). In order to increase computational performance, marginal posterior-based entropies are used for restricting the iterative bi-directional exchange of spatio-temporal information between consecutive images of a time sequence to objects exhibiting a low probability, to be classified correctly according to the HMPM estimation. The Markov models, originally developed for inference on regular graph structures of quadtrees and planar lattices, are adapted to the variable nature of irregular graphs, which are related to information driven image segmentation. Entropy based confidence maps, combined with spatio-temporal relationships of potentially inundated bright scattering vegetation to open water areas, are used for the quantification of the uncertainty in the labeling of each image element in flood possibility masks. With respect to accuracy and computational effort, experiments performed on a bi-temporal TerraSAR-X ScanSAR data-set from the Caprivi region of Namibia during flooding in 2009 and 2010 confirm the effectiveness of integrating hierarchical as well as spatio-temporal context into the labeling process, and of adapting the models to irregular graph structures.

09 17 2010 2240 2258 rs2092240 10.3390/mdpi_crossmark_policy https://creativecommons.org/licenses/by/3.0/ 10.3390/rs2092240 https://www.mdpi.com/2072-4292/2/9/2240 https://www.mdpi.com/2072-4292/2/9/2240/pdf Ormsby Detection of lowland flooding using active microwave systems Photogramm. Eng. Remote Sens. 1985 51 317 Alsdorf Interferometric radar measurements of water level changes on the Amazon flood plain Nature 2000 10.1038/35004560 404 174 Horrit Waterline mapping in flooded vegetation from airborne SAR imagery Remote Sens. Environ. 2003 10.1016/S0034-4257(03)00006-3 85 271 Lu Change detection techniques Int. J. Remote Sens. 2004 10.1080/0143116031000139863 25 2365 Mas Monitoring land-cover changes: A comparison of change detection techniques Int. J. Remote Sens. 1999 10.1080/014311699213659 20 139 Bruzzone Automatic analysis of the difference image for unsupervised change detection IEEE Trans. Geosci. Remote Sens. 2000 10.1109/36.843009 38 1171 10.1109/TGRS.2010.2052816 Martinis, S., Twele, A., and Voigt, S. (2010). Unsupervised extraction of flood-induced backscatter changes in SAR data using Markov image modeling on irregular graphs. IEEE Trans. Geosci. Remote Sens., 48, Accepted. Moser Generalized minimum-error thresholding for unsupervised change detection from SAR amplitude imagery IEEE Trans. Geosci. Remote Sens. 2006 10.1109/TGRS.2006.876288 44 2972 Bruzzone An adaptive semiparametric and context-based approach to unsupervised change detection in multitemporal remote-sensing images IEEE Trans. Image Process. 2002 10.1109/TIP.2002.999678 11 452 Singh Digital change detection techniques using remotely-sensed data Int. J. Remote Sens. 1989 10.1080/01431168908903939 10 989 Melgani A Markov random field approach to spatio-temporal contextual image classification IEEE Trans. Geosci. Remote Sens. 2003 10.1109/TGRS.2003.817269 41 2478 Feitosa Cascade multitemporal classification based on fuzzy Markov chains ISPRS J. Photogramm. Remote Sens. 2009 10.1016/j.isprsjprs.2008.09.011 64 159 Geman Stochastic relaxation, Gibbs distributions and the Bayesian restoration of images IEEE Trans.Pattern Anal. Mach. Intell. 1984 10.1109/TPAMI.1984.4767596 6 721 Li, S.Z. (2009). Markov Random Field Modeling in Image Analysis, Springer-Verlag. [3rd ed.]. Carincotte Unsupervised change detection on SAR images using fuzzy hidden Markov chains IEEE Trans. Geosci. Remote Sens. 2006 10.1109/TGRS.2005.861007 44 432 Benedek Change detection in optical aerial images by a multilayer conditional mixed Markov model IEEE Trans. Geosci. Remote Sens. 2009 10.1109/TGRS.2009.2022633 47 3416 Mota Multitemporal fuzzy classification model based on class transition possibilities ISPRS J. Photogramm. Remote Sens. 2007 10.1016/j.isprsjprs.2007.04.001 62 186 Solberg A Markov random field model for classification of multisource satellite imagery IEEE Trans. Geosci. Remote Sens. 1996 10.1109/36.481897 34 100 Liu Using local transition probability models in Markov random fields for forest change detection Remote Sens. Environ. 2008 10.1016/j.rse.2007.10.002 112 2222 Smits Synthetic aperture radar image segmentation by a detail preserving Markov random field approach IEEE Trans. Geosci. Remote Sens. 1997 10.1109/36.602527 35 844 Jackson Adaptive Bayesian contextual classification based on Markov random fields IEEE Trans. Geosci. Remote Sens. 2002 10.1109/TGRS.2002.805087 40 2454 Kasetkasem An image change detection algorithm based on Markov random field models Canad. J. Remote Sens. 2002 10.1109/TGRS.2002.802498 40 1815 Delignon Unsupervised classification of radar images using hidden Markov chains and hidden Markov random fields IEEE Trans. Geosci. Remote Sens. 2003 10.1109/TGRS.2003.809940 41 675 Kasetkasem Super-resolution land cover mapping using a Markov random field based approach Remote Sens. Environ. 2005 10.1016/j.rse.2005.02.006 96 302 Tso Combining spectral and spatial information into hidden Markov models for unsupervised image classification Int. J. Remote Sens. 2005 10.1080/01431160512331337844 26 2113 Bouman A multiscale image model for Bayesian image segmentation IEEE Trans. Image Process. 1994 10.1109/83.277898 3 162 Heitz Discrete Markov image modeling and inference on the quadtree IEEE Trans. Image Process. 2000 10.1109/83.826777 9 390 Chardin Noniterative manipulation of discrete energy-based models for image analysis Pattern Recog. 2000 10.1016/S0031-3203(99)00073-4 33 573 Cheng Multiscale Bayesian segmentation using a trainable context model IEEE Trans. Image Process. 2001 10.1109/83.913586 10 511 Collet Multiband segmentation based on a hierarchical Markov model Pattern Recognit 2004 10.1016/S0031-3203(04)00190-6 37 2337 Provost Hierarchical Markovian segmentation of multispectral images for the reconstruction of water depth maps Comput. Vis. Image Underst. 2004 10.1016/j.cviu.2003.07.004 93 155 Kato A hierarchical Markov random field model and multi-temperature annealing for parallel image classification Graph. Models Image Process 1996 10.1006/gmip.1996.0002 58 18 Katartzis A hierarchical Markovian model for multiscale region-based classification of vector-valued images IEEE Trans. Geosci. Remote Sens. 2005 10.1109/TGRS.2004.842405 43 548 (2009). eCognition Developer 8 User Guide, Definiens AG. Martinis Towards operational near real-time flood detection using a split-based automatic thresholding procedure on high resolution TerraSAR-X data Nat. Hazards. Earth Syst. Sci. 2009 10.5194/nhess-9-303-2009 9 303 Kittler Minimum error thresholding Pattern Recog. 1986 10.1016/0031-3203(86)90030-0 19 41 Bazi An unsupervised approach based on the generalized Gaussian model to automatic change detection in multitemporal SAR images IEEE Trans. Geosci. Remote Sens. 2005 10.1109/TGRS.2004.842441 43 874 Forney The Viterbi algorithm IEEE Proc. 1973 10.1109/PROC.1973.9030 61 268 Baum A maximization technique occurring in the statistical analysis of probabilistic functions of Markov chains IEEE Ann. Math. Stat. 2005 10.1214/aoms/1177697196 41 164 Swain Bayesian classification in a time-varying environment IEEE Trans. Syst. Man Cybern. 1978 10.1109/TSMC.1978.4309889 8 879 Besag Spatial interaction and the statistical analysis of lattice systems J. R. Stat. Soc. Ser. B 1974 10.1111/j.2517-6161.1974.tb00999.x 36 192 Besag On the statistical analysis of dirty pictures J. R. Stat. Soc. Ser. B 1986 10.1111/j.2517-6161.1986.tb01412.x 48 259 Bruzzone An iterative technique for the detection of land-cover transitions in multispectral remote sensing images IEEE Trans. Geosci. Remote Sens. 1997 10.1109/36.602528 35 858 Bruzzone A neural-statistical approach to multitemporal and multisource remote-sensing image classification IEEE Trans. Geosci. Remote Sens. 1999 10.1109/36.763299 37 1350 Granville Multivariate discriminant analysis and maximum penalized likelihood density estimation J. R. Stat. Soc. Ser. B 1995 10.1111/j.2517-6161.1995.tb02044.x 57 501 Feng Combining belief networks and neural networks for scene segmentation IEEE Trans. Pattern Anal. Mach. Intell. 2002 10.1109/34.993555 24 467 Schumann The utility of spaceborne radar to render flood inundation maps based on multialgorithm ensembles IEEE Trans. Geosci. Remote Sens. 2009 10.1109/TGRS.2009.2017937 47 2801 Richards An explanation of enhanced radar backscattering from flooded forests Int. J. Remote Sens. 1987 10.1080/01431168708954756 8 1093 Hong Evaluation of TerraSAR-X observations for wetland InSAR application IEEE Trans. Geosci. Remote Sens. 2010 10.1109/TGRS.2009.2026895 48 864 Mason Improving river flood extent delineation from synthetic aperture radar using airborne laser altimetry IEEE Trans. Geosci. Remote Sens. 2007 10.1109/TGRS.2007.901032 45 3932