A PROBABILITY WEIGHTED HOUGH TRANSFORM TECHNIQUE FOR SHAPE RETRIEVAL FROM NOISY IMAGERY

被引：3

作者：

COOPER, MA

机构：

来源：

PATTERN RECOGNITION LETTERS | 1994年 / 15卷 / 06期

关键词：

HOUGH TRANSFORM; EDGE DETECTION; IMAGE PROCESSING;

D O I：

10.1016/0167-8655(94)90023-X

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

This paper presents a modification to the Hough transform, which uses probabilistic methods to discriminate between noise and genuine features in the image. The technique is compared with conventional transforms and shown to give a significant enhancement in both signal-to-noise ratio and clutter discrimination.

引用

页码：619 / 630

页数：12

共 10 条

[1]

BOYCE JF, 1987, IMPLEMENTATION HOUGH

[2]

DAVIS LS, 1975, COMPUTER GRAPHICS IM, V4

[3]

DCOOPER MA, 1987, THESIS IMPERIAL COLL

[4] USE OF HOUGH TRANSFORMATION TO DETECT LINES AND CURVES IN PICTURES [J].

DUDA, RO ;

HART, PE .

COMMUNICATIONS OF THE ACM, 1972, 15 (01) :11-&

[5]

HADDON JF, 1987, GENERALISED THRESHOL

[6]

Hough P.V.C., 1962, US Patent, Patent No. 3069654

[7]

JONES GA, 1987, 4TH P SPIE S OPT OPT

[8]

Kittler J., 1983, IMAGE VISION COMPUT, V1, P37, DOI DOI 10.1016/0262-8856(83)90006-9

[9]

OGORMAN F, 1976, IEEE T COMPUT, V25, P449, DOI 10.1109/TC.1976.1674627

[10]

PREWITT JMS, 1972, PICTURE PROCESSING P

← 1 →