›› 2013, Vol. 26 ›› Issue (8): 142-.

• 论文 • 上一篇    下一篇

多云核提取方法在图像分割中的应用

李万臣,田淑娟   

  1. (哈尔滨工程大学 信息与通信工程学院,黑龙江 哈尔滨 150001)
  • 出版日期:2013-08-15 发布日期:2013-09-25
  • 作者简介:李万臣(1963—),男,教授,硕士生导师。研究方向:通信与信息系统。E-mail:liwanchen@hrbeu.edu.cn。田淑娟(1989—),女,硕士研究生。研究方向:图像处理,图像分割,云理论。

Cloudy Nuclear Extract-based Cloud Transformation Method in Image Segmentation

LI Wanchen,TIAN Shujuan   

  1. (College of Information and Communication Engineering,Harbin University of Engineering,Harbin 150001,China)
  • Online:2013-08-15 Published:2013-09-25

摘要:

针对基于云变换的图像分割算法中云模型的构成存在速度慢、计算误差大的问题,提出了一种基于云变换的多云核同时提取的方法,该方法首先将图像灰度信息作出统计,依照一定方式找到所有波峰,并将波峰值作为Ex,最后根据云模型定义构造出正态云模型,并通过实验将云模型与多阈值法分割图像相结合,实验结果证明,该方法不但可以准确快速地计算出云模型、使误差减少、并且使得图像分割效果好、效率高。

关键词: 云模型, 云变换, 云核, 图像分割

Abstract:

In view of the slow speed and great calculation error of the cloud model of the image segmentation algorithm based on cloud transform,this article propose a method based on cloud theory for cloudy nuclear extraction at the same time.The image gray scale information statistics is performed to find all the peaks in accordance with certain pattern.The peak is chosen as an Ex.Finally,according to the definition of cloud models,a normal cloud model is constructed,which is combined through experiments with multi-threshold segmentation images.Experimental results show that this method can not only quickly and accurately calculate the cloud model to reduce the error,but also offer effective image segmentation with good quality.

Key words: cloud model;cloud transform;cloudy nuclea;segmentation

中图分类号: 

  • TP751