USING STATE-SPACE MODELLING IN STRIP KALMAN FILTER

Jozef Racek - Stanislav Marchevský

   This paper describes the application of a strip Kalman filter (KF) to image filtering. Simulation image processing is based on a 3-dimensional (3-D) state-space model with a two-dimensional blurring function and degraded Gaussian white noise. The models of image processing and blurring may be reduced to one by merging their signal flow graphs. The state-space model for the strip filtering can be then derived from it for image restoration. As it can be seen on restored images resulting from the simulation experiment, this 3-D model is very efficient.

Keywords: image processing, Kalman filtering, digital filters, strip filters

[full-paper]