期刊文献+

一种基于形状特征的大规模图像检索方法 被引量:2

A large-scale image retrival based on shape feature
下载PDF
导出
摘要 形状是物体的一个重要属性,在图像检索中发挥着重要作用。本文针对基于形状的大规模图像检索技术进行了研究,提出了一种基于自相似描述符的大规模图像检索方法。自相似描述符是最近提出的一种不同于传统图像形状描述方法的方法,利用图像的重复模式对图像进行描述。本文所述图像检索方法利用自相似描述符作为图像特征,使用近似K-Means方法进行视觉词典生成,之后利用矢量化后的直方图作为图像表示,并使用广义霍夫投票方法进行图像空间验证。在ETHZ图像数据集上的实验结果表明了本文方法的有效性,并具有一定的实用性和推广意义。 Shape is a very important attribute of objects,and it plays an important role in image retrievals.In this paper, we have some research on large-scale image retrival based on shape features, and propose a large-scale image method based on self-similarity descriptor. This method uses self-similarity descriptor as image feature, using the approximate k-means method to generate visual dictionary, to represent the image as a vectorized histogram, and verifies image in Hough space. The experimental in the ETHZ image data set results show the effectiveness of the method.
作者 祝继超 鲁鹏
出处 《软件》 2012年第12期299-304,共6页 Software
关键词 计算机应用技术 形状检索 自相似描述符 视觉词典 霍夫投票算法 Computer Application Technology shape character self-similarity descriptor visual dictionary Hough votingalgorithm
  • 相关文献

参考文献7

  • 1Sivic, J., Zisserman, A. Video Google: A text retrieval approach to object matching in videos[A]. ICCVIC]. 2003. 1470-1477.
  • 2E. Shechtman and M. Irani. Matching local self-similarities across images and videos[A]. IEEE Conference on Computer Vision and Pattern Recognition 2007[C] 2007.
  • 3Ken Chatfield, James Philbin, Andrew Zisserman. Efficient Retrieval of Deformable Shape Classes using Local Self-Similarities[A]. ICCV 2009[C] 2009. 264-271.
  • 4P.O. Hoyer. Non-negative matrix Iactorization with sparseness constraints[A]. JMLR[C] 2004. 1457-1469.
  • 5James Philbin. Scalable Object Retrieval in Very Large Image CoUections[D]. Oxford: University of Oxford, 2010.
  • 6R. Baeza-Yates and B. Ribeiro-Neto. Modem Information Retrieval[R]. Addison-Wesley Long-man Publishing Co., Inc., Boston, MA, USA, 1999.
  • 7V. Ferrafi. ETHZ Shape Classes[OL]. http:// www. vision.ee.ethz.ch/datasets/index, en. html.

同被引文献15

引证文献2

二级引证文献12

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部