基于聚类分析的无监督视觉词典构造方法研究

徐望明; 郑超兵

Chinese

您当前的位置：

首页 >

文章列表页 >

基于聚类分析的无监督视觉词典构造方法研究

更新时间：2026-09-04

- 基于聚类分析的无监督视觉词典构造方法研究
- Unsupervised Visual Dictionary Construction Methods Via Clustering Analysis
- 微电子学与计算机 2016年33卷第3期页码：155-160
- 作者机构：
  
  武汉科技大学机构信息科学与工程学院,湖北,武汉,430081
- 作者简介：
- 基金信息：
  
  国家自然科学基金项目 61105058湖北省教育厅科学技术研究项目 B20121102
- DOI：
  中图分类号：
- 纸质出版：2016
- 稿件说明：
移动端阅览
徐望明, 郑超兵. 基于聚类分析的无监督视觉词典构造方法研究[J]. 微电子学与计算机, 2016, 33(3): 155-160.

XU Wang-ming, ZHENG Chao-bing. Unsupervised Visual Dictionary Construction Methods Via Clustering Analysis[J]. 2016, 33(3): 155-160.
徐望明, 郑超兵. 基于聚类分析的无监督视觉词典构造方法研究[J]. 微电子学与计算机, 2016, 33(3): 155-160. DOI：

XU Wang-ming, ZHENG Chao-bing. Unsupervised Visual Dictionary Construction Methods Via Clustering Analysis[J]. 2016, 33(3): 155-160. DOI：

摘要

视觉词典是运用视觉词袋模型有效表示图像的基础和关键.研究了基于聚类分析的无监督视觉词典构造方法

针对K-Means和层次K-Means(HKM)聚类算法在构造视觉词典时对初始值敏感和容易陷入局部最优的问题进行改进

提出了一种基于谱聚类的视觉词典构造方法

采用对观测数据的相似性矩阵进行特征值分解的方法来实现聚类

可在任意分布的特征空间收敛到全局最优.通过图像检索实验评估了随机采样、K-Means、HKM和改进的谱聚类视觉词典的性能

验证了聚类分析尤其是改进的谱聚类方法构造视觉词典的有效性.

Abstract

Visual dictionary is the basic and key toefficiently represent image with Bag of Visual Words (BoVW) Model. The unsupervised visual dictionary construction methods based on clustering analysis are researched in this paper. Aiming at the problems that K-Means based clustering methods (K-Means and HKM) are sensitive to the initialization and easy to converge to the local optimum

a modified spectral clustering method is proposed to construct visual dictionary

which works in a way to cluster observations via eigenvalue decomposition according to their similarity matrix and can converge to the global optimal regardless of the distribution of feature space. The performance of visual dictionaries constructed by Random Sampling

K-Means

HKM and the improved Spectral Clustering is assessed through image retrieval experiments

and the effectiveness of clustering analysis especially the improved Spectral Clustering method to construct visual dictionary is validated.

关键词

Keywords

references

Sivic J, Zisserman A. Video Google: A text retrieval approach to object matching in videos [C]//Proceeding of the 9th IEEE International Conference on Compute Vision. Nice, France: IEEE, 2003: 1470-1477.

Ng A Y, Jordan M L, Weiss Y. On spectral clustering: analysis and an algorithm [C]// Advances in Neural Information Processing Systems. San Frrancsco: Morgan Kaufmann, 2002: 849-856.

Chatfield K, Lempitsky V, Vedaldi A, et al. The devil is in the details: an evaluation of recent feature encoding methods [C]// Proc BMVC. [s.l.]: IEEE, 2011: 1-12.

UlrikeLuxburg. A tutorial on spectral clustering [J]. Statistics and Computing, 2007, 17(4): 395-416. DOI: 395

Divya Sardana, and Raj Bhatnagar. Graph clustering using mutual K-nearest neighbors [C]// Proceedings of the 10the Intertional Conference AMT. Warsaw, Ploand: Springer, 2014: 35-48.

DavidNister, Henrik Stewenius, Scalable recognition with a vocabulary tree [C]//Proceedings of the 2006 IEEE Computer society Comference on Computer Vistion and Pattern Recognition. Washington, DC USA: ACM, 2006: 2161-2168.

Jinjun Wang, Jianchao Yang, Kai Yu, et al. Locality-constrained linear coding for image classification [C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, San Francisso, CA: IEEE, 2010: 3360-3367.

David G Lowe. Distinctive image features from scale invariant keypoints [J]. International Journal of Computer Vision, 2004, 60(2): 91-110. DOI: 91

Shao H, Svoboda T, Van Gool L. ZUBUD-Zurich buildings database for image based recognition [R]. Swiss Federal Institute of Technology, 2003.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于监控指标关联性的微服务异常检测

基于改进RT-DETR的航拍小目标检测算法

面向处理器分支预测单元的Spectre攻击轻量化防御方法

注意力机制与改进密集卷积引导的信息隐藏模型

基于MSS-YOLO-MPHands融合架构的装配手部动作识别研究