Abstract
Image analysis methods that use histograms defined over non-zero- sized local neighbourhoods have been proposed [1-4]. To better understand such methods, one can study the histograms of infinitesimal neighbourhoods. In this paper we show how the properties of such histograms can be derived through limit arguments. We show that in many cases the properties of these histograms are given by simple expressions in terms of spatial derivatives at the point analyzed.
Cite
CITATION STYLE
Griffin, L. D. (2001). Histograms of infinitesimal neighbourhoods. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 2106, pp. 326–334). Springer Verlag. https://doi.org/10.1007/3-540-47778-0_30
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
