Filtering of Images by Kalman Filter Using VHDL

A Kalman filter is an optimal estimator that infers parameters of interest from indirect inaccurate and uncertain observations. It is recursive so that new measurement can be processed as they arrive. In this paper first we take the input image then add noise to the input image, then give coefficient of noisy image to the VHDL module of kalman filter and finally compare the original input image with filtered image. Kalman filtering method indicates considerable improvement in SNR. The final filtered image shown an approximately 9-dB improvement in SNR with minimal loss of edge information.