In this paper, we have proposed some modifications of the reduced update Kalman filter (RUKF) as applied to filtering of images corrupted by additive noise. First, we have reduced the computational complexity by reducing the state dimensionality. By doing so, it is shown in the paper that the computational requirement is reduced by an order of magnitude while the loss of performance is only marginal. Next, the RUKF is modified using the score function based approach to accommodate the non-Gaussian noise. The image is modeled as a nonstationary mean and stationary variance autoregressive Gaussian process. It is shown in the paper that the stationary variance assumption is reasonable if the nonstationary mean is computed by an edge and detail preserving efficient estimator of local nonstationary mean. Such an estimator called HMSMD filter is also described in the paper. Finally, detailed experimental results are provided which indicate the success of the new filtering scheme.