欢迎访问昆明冶金高等专科学校学报官方网站,今天是 分享到:

昆明冶金高等专科学校学报 ›› 2015, Vol. 31 ›› Issue (3): 60-64.DOI: 10.3969/j.issn.1009-0479.2015.03.012

• 电子信息技术 • 上一篇    下一篇

基于粗集与FCM的快速图像分割的改进方法

张朝全,周绍景2a,梁 颖2b   

  1. 1江西理工大学理学院,江西赣州341000;2昆明冶金高等专科学校a计算机信息学院;b人力资源处,云南昆明650033
  • 收稿日期:2015-04-03 出版日期:2015-07-01 发布日期:2015-07-01
  • 作者简介:张朝全(1979-),男,安徽泗县人,讲师,理学硕士,主要从事数字图像处理、生物特征识别研究。

Improved Methodof Fast Image Segmentation Based on Rough Sets and FCM

ZHANG Chao-quan,ZHOU Shao-jing2a,LIANG Ying2b   

  1. 1Faculty of Science,Jiangxi University of Science and Technology,Ganzhou341000,China;2aFaculty of Computer Information,2bHuman Resources Department,Kunming Metallurgy College,Kunming650033,China
  • Received:2015-04-03 Online:2015-07-01 Published:2015-07-01

摘要:

借鉴神经网络里面竞争学习的思路,通过引入一个抑制因子,来提高最大隶属度的值,对应减小其它隶属度的值,以便达到更快的分类速度,同时实现既突出主要因素,又抑制次要因素的目的。提出了基于粗集与FCM的快速图像分割的改进方法。实验结果表明,该快速图像分割方法,对于被噪声污染的图像有较好的分
割性能,粗集理论在处理不确定性信息方面有着独特的方式和相关信息提取能力,以及和其他智能方法的易融合性,使得粗糙集理论在图像处理领域有良好的应用前景。

关键词: 粗糙集, 图像分割, 模糊聚类

Abstract:

In this paper,the competitivelearning neural network for reference,through the introduction of an inhibitory factor,to maximize the value of membership,the other corresponding decreases the value of membership,in order to achieve faster classification.As to highlight the main factorsand secondary factors,can inhibit the purpose.The improved method of rough set and fast image segmentation based on FCM was put forward.Experimental results show that the fast image segmentation method for image noise pollution has the better segmentation performance,rough set theoryhas a unique way of dealing with uncertain information,the ability to extract relevant information and easy integration of other intelligent methods.Rough set theory has a good application prospect in the field ofimage processing.

Key words: rough sets, image segmentation, fuzzy cluste

中图分类号: