Intrinsic scale space for images on surfaces: The geodesic curvature flow

被引:38
作者
Kimmel, R
机构
[1] Lawrence Berkeley Natl. Laboratory, University of California, Berkeley
来源
GRAPHICAL MODELS AND IMAGE PROCESSING | 1997年 / 59卷 / 05期
关键词
D O I
10.1006/gmip.1997.0442
中图分类号
TP31 [计算机软件];
学科分类号
081202 ; 0835 ;
摘要
A scale space for images painted on surfaces is introduced. Based on the geodesic curvature flow of the iso-gray level contours of an image painted on the given surface, the image is evolved and forms the natural geometric scale space. Its geometrical properties are discussed as well as the intrinsic nature of the proposed flow; i.e., the flow is invariant to the bending of the surface. (C) 1997 Academic Press.
引用
收藏
页码:365 / 372
页数:8
相关论文
共 18 条
[1]   IMAGE SELECTIVE SMOOTHING AND EDGE-DETECTION BY NONLINEAR DIFFUSION .2. [J].
ALVAREZ, L ;
LIONS, PL ;
MOREL, JM .
SIAM JOURNAL ON NUMERICAL ANALYSIS, 1992, 29 (03) :845-866
[2]  
[Anonymous], 1994, GEOMETRIC DRIVEN DIF
[3]  
CHOPP DL, 1993, J EXP MATH, V2, P235
[4]  
EPSTEIN CL, 1987, WAVE MOTION THEORY M
[5]  
FLORACK LMJ, 1994, GEOMETRIC DRIVEN DIF
[6]  
GABOR D, 1965, LAB INVEST, V14, P801
[7]  
GAGE M, 1986, J DIFF GEOM, V23
[8]  
GRAYSON M, 1987, J DIFF GEOM, V26
[9]   SHORTENING EMBEDDED CURVES [J].
GRAYSON, MA .
ANNALS OF MATHEMATICS, 1989, 129 (01) :71-111
[10]  
Kimmel R, 1997, LECT NOTES COMPUT SC, V1252, P236