Measuring 3-D shape similarity using progressive transformations

被引:4
作者
Bribiesca, E [1 ]
机构
[1] UNIV AUTONOMA METROPOLITANA IZTAPALAPA,DEPT MATEMAT,UNIDAD IZTAPALAPA,MEXICO CITY,DF,MEXICO
关键词
measure of 3-D shape similarity; progressive transformations; 3-D object normalization; 3-D shape recognition; volume invariant;
D O I
10.1016/0031-3203(95)00150-6
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
We present a quantitative approach to the measurement of shape similarity among 3-D (three-dimensional) objects. Using voxels, an object is mapped to a representation invariant under translation and rotation. The different objects to be compared are normalized to have the same amount of information (equal number of voxels) and this is termed invariance under volume. When the different objects to be compared are normalized under translation, rotation and volume, a quantity of work (from a physics point of view) is performed that transforms an object O-1 into object O-2 (the transformation of an object into another is performed moving voxels, as if they were bricks). Voxels to move are selected so as to minimize the work involved. The work done by transforming O-1 into O-2 is the measure of dissimilarity between them. Dissimilar objects will have a large quantity of work done to transform one into other, while analogous objects will have a small quantity of work done. When two objects are identical, the quantity of work done is zero. Thus, the distance or shape dissimilarity between two objects can be defined as the amount of work needed to convert one into another. Informally, if two objects to be compared consist of bricks, their shape difference could be ascertained by counting how many bricks we have to move and how far to change one object into another. Copyright (C) 1996 Pattern Recognition Society.
引用
收藏
页码:1117 / 1129
页数:13
相关论文
共 20 条
[1]  
Ballard D.H., 1982, Computer Vision
[2]  
BESL PJ, 1985, ACM COMPUT SURV, V17, P75
[3]  
Boyse JW, 1979, NSF WORKSH REPR 3 DI
[4]   HOW TO DESCRIBE PURE FORM AND HOW TO MEASURE DIFFERENCES IN SHAPES USING SHAPE NUMBERS [J].
BRIBIESCA, E ;
GUZMAN, A .
PATTERN RECOGNITION, 1980, 12 (02) :101-112
[5]   MODEL-BASED 3-DIMENSIONAL INTERPRETATIONS OF TWO-DIMENSIONAL IMAGES [J].
BROOKS, RA .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1983, 5 (02) :140-150
[6]   FROM VOLUMES TO VIEWS - AN APPROACH TO 3-D OBJECT RECOGNITION [J].
DICKINSON, SJ ;
PENTLAND, AP ;
ROSENFELD, A .
CVGIP-IMAGE UNDERSTANDING, 1992, 55 (02) :130-154
[7]  
Guzman A., 1969, Automatic interpretation and classification ofimages
[8]  
Haralick R. M., 1993, COMPUTER ROBOT VISIO, V2
[9]   GLOSSARY OF COMPUTER VISION TERMS [J].
HARALICK, RM ;
SHAPIRO, LG .
PATTERN RECOGNITION, 1991, 24 (01) :69-93
[10]  
HINSKEN L, 1988, INT ARCH PHOTOGRAMME, V27