The task of signal recovery is one of the most important for auto-mated diagnostics and control systems of an energy object. When solv-ing the inverse problems of recovering signals, images and other types of data, spectral distortions and losses occur (in some cases, very sig-nificant ones). They are primarily stipulated due to ill-posedness of these problems, which is the result of loss of information about the original signal due to strong (and even complete) suppression in the observed signal of a part of spectral components, which become indis-tinguishable against the background of errors and noise [1]. Besides, additionalspectral distortions may occur in the process of solving re-covery problems, which depend on specific methods used and their pa-rameters. A method for building a digital correcting filter for pro-cessing the results of solving incorrect inverse problems is proposed, which effectively improves the quality of the solution. The method is based on the use of a singular decomposition of the matrix (SVD) of a system of algebraic equations that approximates the integral operator.