Reduction of Speckle Noise in SAR Image with the Application of DSF Filter
S Varun Alfred Royal, Sagar D Naregal, C Santhosh, Shivani A. Kumar, Mr. Sandeep K V · Zenodo (CERN European Organization for Nuclear Research) · 2019
SAR (Synthetic Aperture Radar) is a mechanism of radar communication crafted for the object reconstruction with the inclusion of both two and three dimension. It is based on the motion of radar antenna over a target region. SAR has an enormous range of practicalities which includes earth observation applications, areas of natural resource management, etc. The former being covering areas in global monitoring, mapping, charting, and land use planning. The latter is its application in fields of forestry agriculture, water quality monitoring and wildlife habitat management. Being so well rounded in its functioning, there is a vital importance in retaining the necessitated information of the SAR imagery during its compression. The presence of inherent speckle noise demeans the quality of SAR. Multiple filters viz. the Lee or Frost filters, based on variation in adaptive filtering coefficient, have been utilised to restore the reflectivity of radar with the underlying assumption of the presence of multiplicative speckle noise in the image. However, due to the occurrence of strong scatters or the structural features (lines or edge) in processing windows, this type of speckle filtering is not as effective as the method proposed by this paper. This is an algorithm for the filtering using Directional Smoothing Filter (DSF). Thus, following noise removal and information preservation, the proposed method helps segue into high data transmission rates and archival ratios.