Loading...
Image segmentation based on graph theory
Author(s)
Cheevasuvit, F.
Date Issued
December 1, 1990
Type
Conference Paper
Abstract
This paper presents an automatic image classification by segmentation method based on graph theory, where each pixel of an image is mapped onto a vertex of graph and the vertices are connected by the link weights. The segmented regions are obtained by agglomerating the vertices which have minimum link weights under a given homogeneity condition. The result shows the segmented regions are connected and the segmented shapes are independent from the problem of scanning direction.
Citation
Proceedings of the International Symposium on Remote Sensing of Environment, 2, 1329-1334, 1990