计算机科学与探索 ›› 2009, Vol. 3 ›› Issue (5): 550-560.DOI: 10.3778/j.issn.1673-9418.2009.05.010

• 学术研究 • 上一篇    

一种主动感知的纹理图像检索方法

曲怀敬1,2+,彭玉华1,吴延荣2,万洪林1,韩 民1   

  1. 1. 山东大学 信息科学与工程学院,济南 250100
    2. 山东建筑大学 信息与电气工程学院,济南 250101
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2009-09-15 发布日期:2009-09-15
  • 通讯作者: 曲怀敬

An Active Perceptual Texture Image Retrieval Method

QU Huaijing1,2+, PENG Yuhua1, WU Yanrong2, WAN Honglin1, HAN Min1   

  1. 1. School of Information Science and Engineering, Shandong University, Jinan 250100, China
    2. School of Information & Electric Engineering, Shandong Jianzhu University, Jinan 250101, China
  • Received:1900-01-01 Revised:1900-01-01 Online:2009-09-15 Published:2009-09-15
  • Contact: QU Huaijing

摘要: 基于Contourlet变换和支持向量机提出了一种新的纹理图像检索方法。在这种方法中,能量和广义高斯分布参数被用做Contourlet子带图像的特征。通过这种表示,提出了由一类和二类支持向量机组成的两阶段检索算法来完成感知相似性测度。通过具有640个纹理图像的VisTex库和具有1 760个纹理图像的Brodatz库证明了所提方法的有效性。实验结果表明,对于这两个纹理库,新的纹理图像检索方法的平均检索率分别达99.38% 和98.07%。

关键词: Contourlet变换, 广义高斯分布, 支持向量机, 相关反馈, 纹理图像检索

Abstract: A new texture image retrieval method based on Contourlet transform and support vector machines (SVMs) is proposed. In this method, the energies and the generalized Gaussian distribution (GGD) parameters are used to represent the Contourlet subband features. Using the representations, a two-run SVM retrieval algorithm which employs an one-class SVM followed by a two-class SVM is proposed to carry out the perceptual similarity measurement. Compared with existing texture image retrieval methods, the proposed retrieval scheme is demonstrated respectively to be effective on the VisTex database of 640 texture images and the Brodatz database of 1 760 texture images. Experimental results have shown that the proposed retrieval scheme can attain 99.38% and 98.07% of the average rates respectively for the two databases.

Key words: Contourlet transform, generalized Gaussian distribution, support vector machine, relevance feedback, texture image retrieval

中图分类号: