Image restoration using a parallel indentification and filtering procedure

In this paper a parallel identification scheme is presented to extend the fast parallel Kalman filter structure of [1] with an on-line identification procedure. It is shown that the identification problem can be specified as a parallel set of one-dimensional autoregressive moving average (ARMA) identification problems. For the case of linear motion blur in the presence of noise a simple and fast parallel identification algorithm is described. Several identification and restoration results are given as examples.