Hybrid image segmentation within a computer vision hierarchy
A generic model of a computer vision system is presented which highlights the critical role of image segmentation. A hybrid segmentation approach, utilising both edge-based and region-based techniques, is proposed for improved quality of segmentation. An image segmentation architecture is outlined and test results are presented and discussed.
Get full access to this article
View all access options for this article.
References
1.
HaralickR. M. and ShapiroL. G., ‘Image segmentation techniques’, Comput. Vision Graphics and Image Processing, 29, pp. 100–132 (1985)
2.
FuK. S. and MuiJ. K., ‘A survey on image segmentation’, Pattern Recognition, 13, pp. 3–16 (1981)
3.
PriceK., ‘Region-based segmentation’, in Encyc. Artif. Intell., ShapiroS. C. (ed.), pp. 877–880 (1987)
PavlidisT. and LiowY., ‘Integrating region growing and edge detection’, IEEE Trans. Patt. Anal, and Mack Intell., 12, pp. 225–233 (1990)
6.
BajscyR., MintzM. and LiebmanE., ‘A common framework for edge detection and region growing’, Univ. Pennsylvania, MS-CIS-86-13, GRASP Lab. Rep. 61 (1986)
7.
GayM., ‘Segmentation using region growing with edges’ Alvey Vision Conf., pp. 115–119 (1989)
8.
CannyJ., ‘Finding edges and lines in images’, MIT AI, Laboratory, Technical Report 720 (June 1983)
