Decision Fusion of Textural Features Derived from Polarimetric Data for Levee Assessment
Cui, M., Prasad, S., Mahrooghy, M., Aanstoos, J.V., Lee, M., & Bruce, L.M. (2012). Decision Fusion of Textural Features Derived from Polarimetric Data for Levee Assessment. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing. IEEE. 5(3), 970-976. DOI:10.1109/JSTARS.2012.2195713.
Texture features derived from Synthetic Aperture Radar (SAR) imagery using grey level co-occurrence matrix (GLCM) can result in very high dimensional feature spaces. Although this high dimensional texture feature space can potentially provide relevant class-specific information for classification, it often also results in over-dimensionality and ill-conditioned statistical formulations. In this work, we propose a GLCM based band grouping followed by a multi-classifier decision fusion framework (MCDF) for a levee health monitoring system that seeks to detect landslides in earthen levees. In this system, texture features derived from the SAR imagery are partitioned into small groups according to different texture features used in GLCM, including energy, correlation, variance, homogeneity (inverse difference moment), entropy and contrast respectively. A multi-classifier system is then applied to each group to perform a local classification. Finally, a decision fusion system is employed to fuse decisions generated by each classifier to make a final decision which is levee or landslide. The resulting system can handle the high dimensionality of the problem very effectively, and only needs a few training samples for training and optimization.