A New Fusion-Based Agricultural Synthetic Aperture Radar Image Despeckling by Using Anisotropic Diffusion and Discrete Wavelet Transform Methods

Abstract

Abstract In agricultural synthetic aperture radar (SAR), the characteristic depression is due to the speckle-noise, which is of a granular form and is multiplicative in nature, leading to common issues as the features of the images get blurred. Precise agricultural SAR images that are needed for despeckling the speckle-noise are obtained by employing despeckling methods. Preserving the fine details of the agricultural SAR images during the process of speckle-noise reduction is a very challenging task for the researcher. Most of the despeckling methods only smooth the edges but do not enhance the edges. The anisotropic diffusion, also known as Perona-Malik diffusion and two-dimensional discrete wavelet transform, is employed to eliminate the speckle-noise and preserve the fine details of the image. The efficiency and robustness of the proposed method are analyzed by its perceptible characteristics and by employing other analyzing parameters like peak-signal-to-noise ratio, structural similarity index measure, universal image quality index, and root mean square error. The robustness and execution time are determined and compared with typical filters and techniques. The proposed method is more robust and has a better ability to be used in realistic applications.