DocumentCode :
2402263
Title :
Shape decomposition by (d1, d2)-weighted skeleton and directional information
Author :
Ablameyko, S.V. ; Frucci, M. ; Marcelli, A.
Author_Institution :
Inst. of Eng. Cybern., Acad. of Sci., Minsk, Byelorussia
Volume :
2
fYear :
1996
fDate :
25-29 Aug 1996
Firstpage :
275
Abstract :
A binary pattern represented by means of its (d1, d2)-weighted skeleton is decomposed into elementary regions of given shape by selecting among its skeletal pixels those representing the considered regions. The selection is achieved by combining distance and directional information associated to each skeletal pixel with the notion of power of expansion of a pixel. A novel algorithm using these concepts to achieve a decomposition into possibly overlapping parts is presented and its performance in terms of naturalness and stability of the achieved decomposition is discussed
Keywords :
image recognition; image segmentation; (d1, d2)-weighted skeleton; binary pattern; directional information; shape decomposition; Bridges; Cybernetics; Diversity reception; Noise figure; Noise measurement; Noise shaping; Object recognition; Shape; Skeleton; Stability;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Pattern Recognition, 1996., Proceedings of the 13th International Conference on
Conference_Location :
Vienna
ISSN :
1051-4651
Print_ISBN :
0-8186-7282-X
Type :
conf
DOI :
10.1109/ICPR.1996.546832
Filename :
546832
Link To Document :
بازگشت