A Recursive Filter for Despeckling SAR Images

Abstract

This correspondence proposes a recursive algorithm for noise reduction in synthetic aperture radar imagery. Excellent despeckling in conjunction with feature preservation is achieved by incorporating a discontinuity-adaptive Markov random field prior within the unscented Kalman filter framework through importance sampling. The performance of this method is demonstrated on both synthetic and real examples.

Synthetic Aperture Radar (Sar) Image Processing Using Hybrid Enhanced Lee Filter With Discrete Wavelet Transform

“…Synthetic Apertures Radar (SAR) technique is popular due to its usability under different weather conditions, its ability to penetrate through clouds and soil [5]. A SAR image is a mean intensity estimate of radar reflectivity of region which is being imaged.…”

“…Where (U,V)denotes pixel location. The multiplicative nature of speckle complicates noise reduction procedure [5].…”

Multi-Feature Based Detection of Landmines Using Ground Penetrating Radar

“…Kalman filter is one of the most frequently used methods for signal processing in various applications and in many cases, it is used for such as removing noise or clutter [44,45] and tracking [46][47][48]. Kalman filter is effective in removing noise or clutter in the GPR signal, which can be modeled as Gaussian noise.…”

Unscented transformation for depth from motion-blur in videos

“…UKF is a recursive state estimation technique which can handle even nonlinear transformations. It has been applied in computer vision for different problems such as image restoration [17] and tracking [20]. To preserve discontinuities in depth, we incorporate an edge-adaptive prior model for the state.…”