I am using the Kalman Stack Filter (
http://rsbweb.nih.gov/ij/plugins/kalman.html) to improve the signal to noise ratio of filtered x-ray signal detected on an imaging plate. Using default values for the variance estimate and bias, the Kalman filter appears to be doing a good job of enhancing the signal. (I have attached a before and after image.) It does, however, add some artifacts (specks) to the final image which were not present in the original image. Applying a median filter seems to remove the specks, but I was wondering where the artifacts came from. (I prefer to avoid additional filtering if not necessary.) I would appreciate hearing from anyone familiar with Kalman filtering.
BTW, I have not tried to change the default variance estimate and bias, since they seem to be appropriate (except for the specks).
Thanks,
Neil
--
ImageJ mailing list:
http://imagej.nih.gov/ij/list.html