A Graph Cut Image Segmentation Based on Quadtree Representation

Yukiya OKUDA  Nobuo SUEMATSU  Akira HAYASHI 

Publication
D - Abstracts of IEICE TRANSACTIONS on Information and Systems (Japanese Edition)  Vol.J91-D  No.10  pp.2540-2549
Publication Date: 2008/10/01
Online ISSN: 1881-0225
Print ISSN: 1880-4535
Type of Manuscript: PAPER
Category: 
Keyword: 
image segmentation,  quadtree,  graph cut, 

Full Text(in Japanese): PDF(994.3KB)

Summary: 
In the graph cut based image segmentation, segmentation problems are interpreted as graph cut problems and solved by one of the existing polynomial time algorithms. Given an image, the existing graph cut segmentation techniques construct a graph whose vertex corresponds to a pixel of the image and hence the graph can be quite large. In this paper, we propose a graph cut segmentation technique which constructs a graph whose vertex corresponds to a region of a quadtree representation of the image. By this technique, the computation time is reduced and moreover, the accuracy of segmentation is improved.