Detection of Anomalies on Earthen Levees with And Without Feature Extraction Using Synthetic Aperture Radar Imagery

Early detection of anomalies on earthen levees by a remote sensing approach could save time and cost versus direct assessment. In this paper, we implemented the support vector machine (svm) supervised classification algorithm with and without feature extraction. Features were extracted using grey level co-occurrence matrix (glcm) features using phase and magnitude imagery of polarimetric synthetic aperture radar (polsar) for the identification of anomalies on levees. The effectiveness of the algorithms is demonstrated using fully quad-polarimetric l-band synthetic aperture radar (sar) imagery from the nasa jet propulsion laboratory's (jpl's) uninhabited aerial vehicle synthetic aperture radar (uavsar). The study area is a section of the lower mississippi river valley in the southern usa, where the us army corps of engineers maintains earthen flood control levees.