AUTOMATIC TARGET SEGMENTATION BY LOCALLY ADAPTIVE IMAGE THRESHOLDING

被引：45

作者：

LIE, WN

机构：

[1] Department of Computer Science and Information Engineering, Yuan-Ze Institute of Technology, Taoyuan

来源：

IEEE TRANSACTIONS ON IMAGE PROCESSING | 1995年 / 4卷 / 07期

关键词：

D O I：

10.1109/83.392347

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

A locally adaptive thresholding algorithm, concerning the extraction of targets from a given field of background, is proposed. Conventional histogram-based or global-type methods are deficient in detecting small targets of possibly low contrast as well. Our research is notable for solving the mentioned problems by introducing 1) shape connectivity measure based on eel-occurrence statistics for threshold evaluation; and 2) no-target identification procedure for modeling a local processing paradigm, In this manner, thresholds are determined adaptively even in the presence of space-varying noise or clutters. Experiments show that our results are reliable and even outperform those that manual operations can achieve for global thresholding.

引用

页码：1036 / 1041

页数：6

共 18 条

[1]

AUBORN JE, 1991, P SOC PHOTO-OPT INS, V1482, P246, DOI 10.1117/12.45700

[2] A NOTE ON THE USE OF THE GRAYLEVEL CO-OCCURRENCE MATRIX IN THRESHOLD SELECTION [J].