Theoretical foundations of spatially-variant mathematical morphology Part I: Binary images

被引:65
作者
Bouaynaya, Nidhal [1 ]
Charif-Chefchaouni, Mohammed [2 ]
Schonfeld, Dan [3 ]
机构
[1] Univ Arkansas, Dept Syst Engn, Donaghey Coll Informat Sci, Little Rock, AR 72204 USA
[2] Inst Natl Postes & Telecommun, Rabat, Morocco
[3] Univ Illinois, Dept Elect & Comp Engn, Chicago, IL 60607 USA
关键词
mathematical morphology; spatially-variant morphology; adaptive morphology; circular morphology; affine morphology; median filter; kernel representation; basis representation; upper-semicontinuity; hit-or-miss transform;
D O I
10.1109/TPAMI.2007.70754
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
We develop a general theory of spatially-variant (SV) mathematical morphology for binary images in the euclidean space. The basic SV morphological operators (that is, SV erosion, SV dilation, SV opening, and SV closing) are defined. We demonstrate the ubiquity of SV morphological operators by providing an SV kernel representation of increasing operators. The latter representation is a generalization of Matheron's representation theorem of increasing and translation-invariant operators. The SV kernel representation is redundant, in the sense that a smaller subset of the SV kernel is sufficient for the representation of increasing operators. We provide sufficient conditions for the existence of the basis representation in terms of upper-semicontinuity in the hit-or-miss topology. The latter basis representation is a generalization of Maragos' basis representation for increasing and translation-invariant operators. Moreover, we investigate the upper-semicontinuity property of the basic SV morphological operators. Several examples are used to demonstrate that the theory of spatially-variant mathematical morphology provides a general framework for the unification of various morphological schemes based on spatially-variant geometrical structuring elements (for example, circular, affine, and motion morphology). Simulation results illustrate the theory of the proposed spatially-variant morphological framework and show its potential power in various image processing applications.
引用
收藏
页码:823 / 836
页数:14
相关论文
共 61 条
[21]  
Fang T, 2003, MACH VISION APPL, V15, P63, DOI 10.1007/s00138-003-0074-1
[22]   FEATURE ENHANCEMENT OF FILM MAMMOGRAMS USING FIXED AND ADAPTIVE NEIGHBORHOODS [J].
GORDON, R ;
RANGAYYAN, RM .
APPLIED OPTICS, 1984, 23 (04) :560-564
[23]   Nonlinear multiresolution signal decomposition schemes - Part I: Morphological pyramids [J].
Goutsias, J ;
Heijmans, HJAM .
IEEE TRANSACTIONS ON IMAGE PROCESSING, 2000, 9 (11) :1862-1876
[24]   MORPHOLOGICAL REPRESENTATION OF DISCRETE AND BINARY IMAGES [J].
GOUTSIAS, J ;
SCHONFELD, D .
IEEE TRANSACTIONS ON SIGNAL PROCESSING, 1991, 39 (06) :1369-1379
[25]  
HANBURY A, 2002, VISION NONTRADITIONA
[26]   Minimal-memory bit-vector architecture for computational mathematical morphology using subspace projections [J].
Handley, JC .
IEEE TRANSACTIONS ON IMAGE PROCESSING, 2005, 14 (08) :1088-1095
[27]  
HEIJMANS HJ, 1991, COMPUT VISION GRAPH, V54, P74
[28]   Nonlinear multiresolution signal decomposition schemes - Part II: morphological wavelets [J].
Heijmans, HJAM ;
Goutsias, J .
IEEE TRANSACTIONS ON IMAGE PROCESSING, 2000, 9 (11) :1897-1913
[29]   THE ALGEBRAIC BASIS OF MATHEMATICAL MORPHOLOGY .1. DILATIONS AND EROSIONS [J].
HEIJMANS, HJAM ;
RONSE, C .
COMPUTER VISION GRAPHICS AND IMAGE PROCESSING, 1990, 50 (03) :245-295
[30]  
Heijmans HJAM., 1994, MORPHOLOGICAL IMAGE