Performance Evaluation of Speckle Noise Reduction in SAR Image using DSF Filter

Anshita Shrivastava; Kranti Jain; Prashant Richarya

CALL FOR PAPERS : Feb-2026

Submission Last Date

25-Feb-26

Submit Manuscript Online

FOR AUTHORS

FOR REVIEWERS

ARCHIEVES

DOWNLOADS

Open Access

Performance Evaluation of Speckle Noise Reduction in SAR Image using DSF Filter

Author(s):

Anshita Shrivastava , Chhatrapati Shivaji Institute of Technology, Durg; Kranti Jain, Chhatrapati Shivaji Institute of Technology, Durg; Prashant Richarya, Chhatrapati Shivaji Institute of Technology, Durg

Keywords:

Directional Smoothing Filter (DSF), Artificial Intelligence (AI), Electromagnetic (EM), Synthetic Aperture Radar (SAR), Shuttle Imaging Radar (SIR)

Abstract:

Speckle Noise filters tries to restore the reflectivity of radar assuming that multiplicative speckle noise id present in the image. Some of the best known filters, namely the Lee filter, Kuan filters or Frost, are adaptive filters which are based on the local statistics that is computed within a fixed size square window. Therefore, the speckle is reduced as a function of heterogeneity that is measured by the local coefficient of variation. As the radar reflectivity suffer significant variations due to the occurrence of strong scatters or the structural features (lines or edge) in processing windows, this type of speckle filtering is not so effective. This paper presents an algorithm for the filtering using Directional Smoothing Filter. SAR images can be used in a myriad of earth observation applications covering areas in global monitoring, mapping, charting, and land use planning. There is also the area of natural resource management, including forestry agriculture, water quality monitoring and wildlife habitat management. With such a diverse set of applications a wide variety of system requirements: the challenge of including SAR image compression in a system is to preserve sufficient information content of the imagery, while providing better noise removal from the SAR image. Thus by removing noise and by preserving the information the proposed method allow high data transmission rates and archival ratios.