THE USE OF PERCEPTUAL ORGANIZATION IN THE PREDICTION OF GEOMETRIC STRUCTURES

被引:4
作者
DENASI, S
QUAGLIA, G
RINAUDI, D
机构
[1] Istituto Elettrotecnico Nazionale Galileo Ferraris, 10135 Torino
关键词
FEATURE EVIDENCE; PERCEPTUAL ORGANIZATION; GEOMETRICAL GROUPING;
D O I
10.1016/0167-8655(92)90071-7
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
A strategy is proposed for detecting geometric structures by perceptual organization, without prior knowledge of the image contents. A saliency criterion has been defined, that involves the length and the luminance contrast of the edges of an image. Starting from the most salient edges, perceptual organization guides the prediction of the geometric structures that are suggested by parallelism and connectivity relationships. This strategy allows an efficient analysis of the large amount of image features, pointing out at first the main structures present in the scene.
引用
收藏
页码:529 / 539
页数:11
相关论文
共 9 条
[1]   CONTOUR CODING FOR IMAGE DESCRIPTION [J].
GRATTONI, P ;
GUIDUCCI, A .
PATTERN RECOGNITION LETTERS, 1990, 11 (02) :95-105
[2]   PICTORIAL RECOGNITION AS AN UNLEARNED ABILITY - A STUDY OF 1 CHILDS PERFORMANCE [J].
HOCHBERG, J ;
BROOKS, V .
AMERICAN JOURNAL OF PSYCHOLOGY, 1962, 75 (04) :624-&
[3]  
Kennedy J., 1974, PSYCHOL PICTURE PERC
[4]  
Koffka Kurt, 1963, PRINCIPLES GESTALT P
[5]   3-DIMENSIONAL OBJECT RECOGNITION FROM SINGLE TWO-DIMENSIONAL IMAGES [J].
LOWE, DG .
ARTIFICIAL INTELLIGENCE, 1987, 31 (03) :355-395
[6]   USING PERCEPTUAL ORGANIZATION TO EXTRACT 3-D STRUCTURES [J].
MOHAN, R ;
NEVATIA, R .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1989, 11 (11) :1121-1139
[7]  
Rock I., 1983, LOGIC PERCEPTION
[8]  
Witkin A. P., 1983, HUMAN MACHINE VISION, V1, P481
[9]  
[No title captured]