Image restoration using second-order Kalman filter

The purpose of this paper was to suggest a new restoration technique for images with additive noise. The proposed technique, a one-dimensional second-order model, is designed on the basis of the second-order Kalman filter system, which is an extension from the conventional first-order Kalman model. The proposed technique treats the gray value of a pixel on each line of the image in a way highly correlated to two previously processed pixels. While the first-order Kalman filter is feasible for the image restoration problem, improved quality of restoration can be obtained if further information from the neighbored pixels is available. Three sets of different images were tested to verify the suitability of the second-order Kalman filter for image restoration. Experimental results demonstrate that the proposed technique provides better restoration quality than the first-order model does.