I. Numata and S. Komatsu: Improved Blind Deconvolution Algorithm Using a Sub-Band Filter Bank , Technical Report (Advanced Research Institute for Science and Engineering, Waseda Univ.), No.2001-4 (2001), 1-11.

The performance of the Fourier iterative algorithm for blind deconvolution is improved by adopting a sub-band filter bank. In the proposed method a blurred image is first decomposed into its sub-band images by the corresponding decomposition sequences, then each of the sub-band images is blind-deconvolved in parallel. Finally the deconvolved sub-band images are added up again to reconstruct the object. The low frequency sub-band image is blind-deconvolved with the Fourier iterative blind deconvolution algorithm proposed by Ayers and Dainty. The others are blind-deconvolved with simulated annealing algorithm. The feasibility of the proposed method is examined on computer simulation.