Synthetic Aperture Radar (SAR) Image Processing using Hybrid Enhanced Lee Filter with Discrete Wavelet Transform

Abstract

Synthetic Aperture Radar (SAR) images are significant application of optical satellite images due to its ability
to work in any weather situations. This has applications in ecology, geology, agriculture, hydrology, oceanography and
military so on. With enhancement of SAR technology, resolution of images has increased. These images are corrupted by
a strong noise, known as Speckle noise. This noise reduces overall quality of image. So it is necessary to erase speckle
noise and preserve all features in SAR images with suitable strategy. In this paper, we proposed a new hybrid technique
that combines Wavelet based demonizing and anisotropic diffusion filter, both these techniques provides better filtering
capability and enhances edges. By applying these techniques the efficiency of system is increased and noise is reduced to
the greater extent. To Remove the Speckle Noise efficiently a New hybrid technique is developed using Wavelet based
anisotropic filter which works efficiently and removes the full removal of Speckle and preserve the edges up to greater
extent. The Results are evaluated in terms of statistical parameters and compared with existing algorithm. The
performance is evaluated in terms of COC, ENL and PSNR, the proposed hybrid approach provide fully demonized
images when compared with existing model.