Enhanced Wiener Restoration of Images Based on the Haar Wavelet Transform

This paper proposes two techniques for the enhancement of the performance of the Linear Minimum Mean Square Error (LMMSE) or Wiener restoration algorithm. The first approach relies on adaptively merging the Wiener restoration algorithm with a regularized restoration algorithm that is implemented iteratively. The merging process allows the capturing of edges regions from the Wiener restoration result and the capturing of flat areas from the regularized restoration result. The second approach is a smoothing technique to reduce noise in flat areas in the restored images. The decision that is used for selecting the regions in the restored image, in which either the regularization or the smoothing algorithm can be used, is dependent on the 2D Haar wavelet transform.