Spectral cluster ensemble image segmentation based on similarity matrix

ZHANG Qi; LU Zhi-mao; XU Sen; LIU Chen; SUI Yi

doi:10.13873/j.1000-97872013.10.031

Chinese

您当前的位置：

首页 >

文章列表页 >

Spectral cluster ensemble image segmentation based on similarity matrix

更新时间：2025-09-11

- Spectral cluster ensemble image segmentation based on similarity matrix
- Vol. 32, Issue 10, Pages: 21-23+26(2013)
- 作者机构：
  
  1. 哈尔滨工程大学信息与通信工程学院
  2. 盐城工学院信息工程学院
  3. 哈尔滨师范大学教育科学学院
- 作者简介：
- 基金信息：
- DOI：10.13873/j.1000-97872013.10.031
  CLC： TP391.41
- Published：2013
- 稿件说明：
移动端阅览
[1]张琦,卢志茂,徐森,刘晨,隋毅.基于相似度矩阵的谱聚类集成图像分割[J].传感器与微系统,2013,32(10):21-23+26.

ZHANG Qi, LU Zhi-mao, XU Sen, et al. Spectral cluster ensemble image segmentation based on similarity matrix[J]. 2013, 32(10): 21-23+26.
[1]张琦,卢志茂,徐森,刘晨,隋毅.基于相似度矩阵的谱聚类集成图像分割[J].传感器与微系统,2013,32(10):21-23+26. DOI： 10.13873/j.1000-97872013.10.031.

ZHANG Qi, LU Zhi-mao, XU Sen, et al. Spectral cluster ensemble image segmentation based on similarity matrix[J]. 2013, 32(10): 21-23+26. DOI： 10.13873/j.1000-97872013.10.031.

摘要

针对谱聚类集成算法计算复杂度高

难以应用到大规模图像分割处理的问题

提出一种将MS和基于超边相似度矩阵的谱聚类集成算法（HSMCESA）相结合的彩色图像分割算法（MS-HSMCESA）。首先

采用MS算法对彩色图像进行预分割

计算分割得到的每个区域的所有像素的彩色向量的平均值

以此作为HSMCESA的输入。在HSMCESA的谱分解过程中

通过矩阵变换对特征值分解进行近似求解

大大降低了算法的时间复杂度。对比实验表明:MS-HSMCESA较MS-Kmeans和MS-Ncut算法能获得更好的分割质量。

Abstract

Aiming at problem that spectral cluster ensemble algorithm is hard to be applied in large scale image segmentation processing because of high computational complexity

a new color image segmentation method combining mean shift（MS） and Hyperedges ’similarity matrix-based custer ensemble spectral algorithm（HSMCESA） named MS-HSMCESA is proposed.First

some regions are obtained through pre-segmentation by MS algorithm.The average value of color vectors in each region are considered as input of HSMCESA.Through matrix transformation

it computes eigenvalues of a small matrix to obtain the eigenvalues of the similarity matrix to reduce the time complexity.Experimental results show that MS-HSMCESA can always obtained better image segmentation quality than MS-Kmeans and MS-Ncut algorithm.

关键词

Keywords

references

使用谱聚类算法解决文本聚类集成问题 [J]. 徐森,卢志茂,顾国昌. 通信学报 . 2010(06)

Views

176

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Improvements of acceleration k-means based spectral clustering algorithm for image segmentation

Character paper-cutting design based on convolutional neural network

Blood serum level detection in blood-collecting vessel based on image recognition

Color image segmentation based on disjunctive normal level set

A pore detection algorithm based on joint feature constraint

Related Author

LI Chang-xing

HUANG Yan-hu

ZHI Xiao-bin

XIE Xiao-juan

YAO Chao

LIU Guihua

ZHAO Sen

LIN Jie

Related Institution

School of Communication and Information Engineering,Xi’an University of Posts and Telecommunication

School of Science,Xi’an University of Posts and Telecommunications

College of Public Administration,University of Electronic Science and Technology of China

Sichuan Key Laboratory of Special Environment Robot Technology

College of Information Engineering,Southwest University of Science and Technology

⁰