In Section III, page 184, colour images are mapped to the HSI (hue,saturation, intensity) colour space. After that, the HSI image is filtered using the Volterra filter in Matlab.
The mathematics underlying Volterra filtering is given in
J. August, Volterra filtering of noisy images of curves, Proc. ECCV 2002:
A very detailed explanation of the Volterra filtering method is given in
J. August, Biased curve indicator random field filters for enhancement of contours in images, 2003:
http://www.google.ca/patents/US20030118246
Another place to look is
T. Koh, E.J. Powers, Second-order Volterra filtering and its application to nonlinear system identification, IEEE Tans. on Acoustics, Speech and Signal Proc. 33, 1985, no. 6, 1445-1455:
In Section III, page 184, colour images are mapped to the HSI (hue,saturation, intensity) colour space. After that, the HSI image is filtered using the Volterra filter in Matlab.
The mathematics underlying Volterra filtering is given in
J. August, Volterra filtering of noisy images of curves, Proc. ECCV 2002:
A very detailed explanation of the Volterra filtering method is given in
J. August, Biased curve indicator random field filters for enhancement of contours in images, 2003:
http://www.google.ca/patents/US20030118246
Another place to look is
T. Koh, E.J. Powers, Second-order Volterra filtering and its application to nonlinear system identification, IEEE Tans. on Acoustics, Speech and Signal Proc. 33, 1985, no. 6, 1445-1455: