Two Dimensional Recursive Estimation with Application to Real-Time Filtering of Images.
John W. Woods · 1978
The main goal of this program is the development of efficient methods for the two-dimensional (2-D), recursive filtering of images corrupted by noise. This goal is to be achieved by the use of our extension of Kalman recursive estimation techniques to two-dimensions, which permits order-of-magnitude savings in computer time and storage over previously developed techniques for optimal recursive estimation of noisy images. Such images could arise for example in a communications context by an image transmission system which is noise limited because of power constraints. (Author)