Variances, covariances and artifacts in image deconvolution

In this paper we investigate the role of estimator variance and covariance in signal and image deconvolution, where artificial structures and large fluctuations are frequently observed. The deconvolution procedure adopted in this paper is based on the Least Squares Method and the regularization, on a minimum variance property. The whole covariance matrices were taken into account in every step of the deconvolution and regularization procedures. We show that the covariances created by the deconvolution procedure are at the origin of the observed artifacts in the recovered signal.