DIRECTIONAL STATISTICS IN NONLINEAR VECTOR FIELD FILTERING

被引:8
作者
NIKOLAIDIS, N [1 ]
PITAS, I [1 ]
机构
[1] UNIV THESSALONIKI,DEPT ELECT ENGN,GR-54006 THESSALONIKI,GREECE
关键词
ANGULAR STATISTICS; MEDIAN FILTERS; COLOR IMAGE PROCESSING; VECTOR FIELD FILTERING;
D O I
10.1016/0165-1684(94)90151-1
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
Vector fields play an important role in the area of color image processing and image sequence processing. In certain cases (e.g. motion direction estimation, Hue color component filtering), vector representation and manipulation in a polar coordinate system can be advantageous. In this paper we present a review of the theory of angular statistics along with some new theoretical results. Emphasis is given to direction estimators. An application illustrating the use of measures of angular dispersion in color edge detection is provided. Various ordering principles for directional data are presented. The effectiveness of direction estimators in filtering of directional data is studied in simulations involving color images and velocity fields.
引用
收藏
页码:299 / 316
页数:18
相关论文
共 20 条
[1]  
Astola, Haavisto, Neuvo, Vector median filters, Proc. IEEE, 78, pp. 678-690, (1990)
[2]  
Batschelet, Circular Statistics in Biology, (1981)
[3]  
Foley, van Dam, Feiner, Hughes, Computer Graphics: Principles and Practice, (1990)
[4]  
Gould, A regression technique for angular variates, Biometrics, 25, pp. 683-700, (1969)
[5]  
Hardie, Arce, Ranking in R<sup>p</sup> and its use in multivariate image estimation, IEEE Trans. Circuits and Systems for Video Technology, 1 CSVT, 2, pp. 197-209, (1991)
[6]  
Jain, Fundamentals of Digital Image Processing, (1989)
[7]  
Kotropoulos, Pitas, Multichannel L-filters based on M-ordering, IEEE Workshop on Nonlinear Digital Signal Processing, (1993)
[8]  
Lenth, Robust measures of location for directional data, Technometrics, 23, 1, pp. 77-81, (1981)
[9]  
Liu, On a notion of data depth based on random simplices, The Annals of Statistics, 18, pp. 405-414, (1990)
[10]  
Liu, Singh, Ordering directional data Concepts of data depth on circles and spheres, The Annals of Statistics, 20, 3, pp. 1468-1484, (1992)