Restoration of noisy images blurred by a random point spread function

An estimation method based on the constrained least-squares principle is presented for the restoration of images distorted by a random point spread function and additive measurement noise. The proposed filter modifies the conventional constrained least-squares filter by incorporating additional statistical characteristics about the randomness of the point spread function. Simulation results show that the proposed method outperforms the conventional constrained least-squares method, which neglects the randomness of the point spread function. For space-invariant systems, the modified constrained least-squares filter can be constructed in the discrete frequency domain and its overall computation can be carried out using the fast Fourier transform