2020 · 13 citations · 13 references
Natural SciencesEarth SciencesCrop SensingDigital ImagingRemote SensingImage FilteringLand Cover ClassificationRadar Signal ProcessingRemote Sensing ImageryAdaptive FiltersFilteringFrequency Domain ProcessingLand Cover Classification SchemesRadar-based Crop ClassificationSpatial Domain EnhancementSpeckle Noise ReductionWavelet Transform
Synthetic Aperture Radar (SAR) produces high resolution, day & night and weather independent images which provide crucial qualitative and quantitative information required for performing various important remote sensing applications like land cover classification, crop classification, etc. But these images get inherently corrupted with the multiplicative speckle noise due to the coherent nature of SAR which significantly influences the information extracted from these images and thus enhances the error rates of the classification algorithms. To overcome these shortcomings, several authors have proposed various wavelet domain or frequency domain or spatial domain based adaptive or non-adaptive despeckling filters, In this work, a comprehensive survey is conducted on the filtering strategies adapted by Frost filter which is one of the most popular adaptive despeckling filter proposed till date and its proposed variants. The methodologies, strengths and limitations of different models of the Frost filter proposed over the years in chronological order to show how their filtering strategies have evolved with time is discussed. This survey also includes comparative qualitative and quantitative analyses of the outputs obtained by applying them on SAR and Polarimetric SAR images. Finally, our survey is conclude by summarizing the unexplored challenges in this field which went unnoticed while designing these filters.
13
A tutorial on synthetic aperture radar
Alberto Moreira, Pau Prats, Marwan Younis et al. · IEEE Geoscience and Remote Sensing Magazine · 2013 · 2.7K citations