No CrossRef data available.
Published online by Cambridge University Press: 21 December 2007
This paper presents a set of shape parameters giving an estimation of features for 3D objects, in particular bounded surfaces such as molecular island of self-assembled carbon chains on silicon, in order to classify them. These features are quantified both in 2D and 3D and concern an evaluation of the circularity degree or the fractal degree of the object boundary in 2D and a study of peak dispersion and height variations in 3D. The originality of this approach mainly lies in the way of estimating the isotropic and fractal degrees from the boundary of the surface and in the combination of different data, such as height or distance to the edges, for peak characterization. An application to small molecular islands of carbon, studied in nano-elec-tronics, is also presented, in order to illustrate the purpose and the efficiency of the previous shape parameters. This application also includes a new method of adaptive filtering, processed from distance maps.