Blind Image Deconvolution: An Algorithmic Approach to Practical Image Restoration

The goal of image restoration is to reconstruct the original scene from a degraded observation. This recovery process is critical to many image processing applications. Although classical linear image restoration has been thoroughly studied [1], [2], the more di cult problem of blind image restoration has numerous research possibilities. Blind image restoration is the process of estimating both the true image and the blur from the degraded image characteristics using partial information about the imaging system. In classical linear image restoration, the blurring function is given, and the degradation process is inverted using one of the many known restoration algorithms. The various approaches that have appeared in literature depend upon the particular degradation and image models [1], [2]. In many imaging applications an observed image g(x; y), neglecting additive noise, can be estimated to be the two-dimensional convolution of the true image f(x; y) with a linear shiftinvariant blur, also known as the point-spread function (PSF), h(x; y). That is,