Conventional spatially-adaptive regularized image restoration schemes weight the amount of regularization according to the spatial content of an image. In this correspondence, we first separately decorrelate the signals under analysis into uncorrelated components and then weight the amount of regularization performed to these components accordingly. The proposed approach works better than conventional schemes especially in edge regions. * Corresponding author This paper is a postprint of a

more »

... submitted to and accepted for publication in IEE Proceedings-Vision, Image and Signal Processing and is subject to Institution of Engineering and Technology