Semi-blind deconvolution of defocused image with MCMA

In this paper, a new approach for blurred noisy image restoration is presented. The blurred noisy image is assumed to be the output of a linear space-invariant system with an unknown point spread function contaminated by an additive noise. The scheme passes the blurred noisy image through a two-dimensional finite impulse response filter whose parameters are updated by the modified constant modulus algorithm with averaging and variable step size. When convergence occurs, the output of the filter is an estimate of the unobserved true image. Experimental results show that the proposed scheme is effective.