The aim of the paper is to put forward an iterative procedure for solving a class of ill-posed inverse optical problems, namely those that are expressed by a convolution-type Fredholm equation of the first kind. This class includes the problem of recovering a signal distorted by a system and in particular the object restoration problem. The method proposed is based on changing the operator, i.e. on converting the Fredholm equation of the first kind into one of second kind. An iterative algorithm for implementing the method is suggested and compared with other iterative procedures.