Relations between anisotropic diffusion and robust statistics are desc
ribed in this paper, Specifically, we show that anisotropic diffusion
can be seen as a robust estimation procedure that estimates a piecewis
e smooth image from a noisy input image, The ''edge-stopping'' functio
n in the anisotropic diffusion equation is closely related to the erro
r norm and influence function in the robust estimation framework, This
connection leads to a new ''edge-stopping'' function based an Tukey's
biweight robust estimator that preserves sharper boundaries than prev
ious formulations and improves the automatic stopping of the diffusion
. The robust statistical interpretation also provides a means for dete
cting the boundaries (edges) between the piecewise smooth regions in a
n image that has been smoothed with anisotropic diffusion, Additionall
y, we derive a relationship between anisotropic diffusion and regulari
zation with Line processes. Adding constraints on the spatial organiza
tion of the line processes allows us to develop new anisotropic diffus
ion equations that result in a qualitative improvement in the continui
ty of edges.