In this paper optimal rates of convergence are derived for estimates o
f sets in N-dimensional ''black and white'' pictures under smoothness
conditions. It is assumed that the boundaries of the ''black'' regions
have a smooth parameterisation, that is, that the boundaries are give
n by smooth functions from the sphere S-N-1 into R(N). Furthermore, cl
asses of convex regions are considered. Two models are studied: edge e
stimation models motivated by image segmentation problems and density
support estimation.