AUTOMATIC THRESHOLD SELECTION USING HISTOGRAMS BASED ON THE COUNT OF 4-CONNECTED REGIONS

被引：7

作者：

LUIJENDIJK, H

机构：

[1] Philips Medical Systems, 5680 DA Best

来源：

PATTERN RECOGNITION LETTERS | 1991年 / 12卷 / 04期

关键词：

IMAGE PROCESSING; AUTOMATIC THRESHOLD SELECTION; IMAGE SEGMENTATION; HISTOGRAM METHOD; RANGE IMAGE; PERCOLATION THEORY;

D O I：

10.1016/0167-8655(91)90035-K

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

For automatic threshold selection, the use of two histograms is proposed, showing the number of 4-connected black and white regions respectively, as a function of the applied threshold level. The major advantages are the smoothness of the histogram envelope and the fixed relation between the peak height of the histogram and the corresponding area in the image.

引用

页码：219 / 228

页数：10

共 6 条

[1]

JOHANNSEN G, 1982, 6TH P INT C PATT REC, P140

[2] THRESHOLD SELECTION METHOD FROM GRAY-LEVEL HISTOGRAMS [J].

OTSU, N .

IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS, 1979, 9 (01) :62-66

[3] A SURVEY OF THRESHOLDING TECHNIQUES [J].

SAHOO, PK ;

SOLTANI, S ;

WONG, AKC ;

CHEN, YC .

COMPUTER VISION GRAPHICS AND IMAGE PROCESSING, 1988, 41 (02) :233-260

[4] THRESHOLD SELECTION TECHNIQUE [J].

WESZKA, JS ;

NAGEL, RN ;

ROSENFELD, A .

IEEE TRANSACTIONS ON COMPUTERS, 1974, C 23 (12) :1322-1326

[5] THRESHOLD SELECTION USING QUADTREES [J].

WU, AY ;

HONG, TH ;

ROSENFELD, A .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1982, 4 (01) :90-94

[6]

Zallen R., 1983, PHYSICS AMORPHOUS SO, P135

← 1 →