We indicate how granulometries may be useful in the analysis of random sets. After defining a size distribution function which may be used as a summary statistic in exploratory data analysis, we propose a Hanisch-type estimator and construct new Markov random set models which favour certain sizes above others. The models are illustrated by simulated realisations.
|Title of host publication||Mathematical morphology and its applications to image and signal processing|
|Editors||H.J.A.M. Heijmans, J.B.T.M. Roerdink|
|Place of Publication||Dordrecht|
|Publisher||Kluwer Academic Publishers|
|Publication status||Published - 1998|
|Name||Computational imaging and vision|