Unsupervised segmentation of synthetic aperture radar inundation imagery using the level set method

In this paper, we proposed an unsupervised algorithm to identify the flooded areas from synthetic aperture radar (SAR) images based on texture information derived from the gray-level co-occurrence matrices (GLCM) texture analysis. Here, five GLCM features, namely, energy, contrast, homogeneity, correlation and entropy, are extracted from a SAR image. These features are input to an image segmentation algorithm using a level set method to identify flooded and dry areas. Experiments were conducted on the RADARSAT-2 images of severely flooded areas near Chaopraya rivers, Thailand, in 2011, for which contemporaneous ground data exists for validation. Our results demonstrate that the proposed algorithm is able to successfully segment various flood regions and achieve improvement over existing published unsupervised algorithms.

[1]  D. Mason,et al.  Flood boundary delineation from Synthetic Aperture Radar imagery using a statistical active contour model , 2001 .

[2]  S. Osher,et al.  Algorithms Based on Hamilton-Jacobi Formulations , 1988 .

[3]  Sergei Vassilvitskii,et al.  k-means++: the advantages of careful seeding , 2007, SODA '07.

[4]  Stephen V. Stehman,et al.  Selecting and interpreting measures of thematic classification accuracy , 1997 .

[5]  E. Nezry,et al.  Structure detection and statistical adaptive speckle filtering in SAR images , 1993 .

[6]  Pierre Kornprobst,et al.  Mathematical problems in image processing - partial differential equations and the calculus of variations , 2010, Applied mathematical sciences.

[7]  Paul D. Bates,et al.  Flood Detection in Urban Areas Using TerraSAR-X , 2010, IEEE Transactions on Geoscience and Remote Sensing.

[8]  Chao Yang,et al.  Flood detection and mapping of the Thailand Central plain using RADARSAT and MODIS under a sensor web environment , 2012, Int. J. Appl. Earth Obs. Geoinformation.

[9]  Anil K. Jain,et al.  Texture Analysis , 2018, Handbook of Image Processing and Computer Vision.

[10]  Shahid Habib,et al.  Satellite Remote Sensing and Hydrologic Modeling for Flood Inundation Mapping in Lake Victoria Basin: Implications for Hydrologic Prediction in Ungauged Basins , 2009, IEEE Transactions on Geoscience and Remote Sensing.

[11]  Russell G. Congalton,et al.  Assessing the accuracy of remotely sensed data : principles and practices , 1998 .

[12]  Margarida Silveira,et al.  Separation Between Water and Land in SAR Images Using Region-Based Level Sets , 2009, IEEE Geoscience and Remote Sensing Letters.

[13]  J. Sethian,et al.  Fronts propagating with curvature-dependent speed: algorithms based on Hamilton-Jacobi formulations , 1988 .

[14]  V. Mani,et al.  Multi-Temporal Satellite Imagery for Flood Damage Assessment , 2013 .

[15]  Robert M. Haralick,et al.  Textural Features for Image Classification , 1973, IEEE Trans. Syst. Man Cybern..

[16]  David A. Clausi,et al.  Unsupervised segmentation of synthetic aperture Radar sea ice imagery using a novel Markov random field model , 2005, IEEE Transactions on Geoscience and Remote Sensing.

[17]  N. Otsu.  A threshold selection method from gray level histograms , 1979 .

[18]  Paul D. Bates,et al.  The Utility of Spaceborne Radar to Render Flood Inundation Maps Based on Multialgorithm Ensembles , 2009, IEEE Transactions on Geoscience and Remote Sensing.

[19]  John G. Daugman,et al.  Complete discrete 2-D Gabor transforms by neural networks for image analysis and compression , 1988, IEEE Trans. Acoust. Speech Signal Process..

[20]  Chunming Li,et al.  Level set evolution without re-initialization: a new variational formulation , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).