2-D Kalman Filtering For Image Restoration In The Presence Of Additive And Signal-dependent Degradations
M.R. Azimi-Sadjadi, E. Nebot · 2005
This paper considers the problem of image restoration in the presence of multiple degradations. A new 2-D Kalman filtering procedure is introduced which takes into account the effects of multiplicative signal-dependent noise, additive white Gaussian (WG) noise and linear shift invariant (LSI) blur. The effectiveness of the proposed method is examined on synthetic aperture radar (SAR) imagery that contain multiplicative speckle noise.