期刊文献+

一种新型的基于神经网络的图像检索算法 被引量:2

A Novel Method of Neural-Network-Based Image Retrieval Using Multiple Features
下载PDF
导出
摘要 分析传统基于多特征的图像检索算法,针对于传统算法核心仅放在如何更多的提取图像的底层特征,却忽视了如何组织这些特征,提出一种改进的基于神经网络的图像检索算法。即在深入提取图像颜色以及形状特征的同时,将图像检索中特征权值调整问题转化为最优化问题,进一步利用神经网络求取权重的最优解,使检索效果达到最优。仿真实验表明,该算法能够准确和高效地查找出所需的图像,具有很强的鲁棒性和检索能力。 According to the analysis of traditional arithmetic of multi-character image retrieval, we propose a novel arithmetic of image retrieval based on neural network because of the traditional arithmetic attach importance to the feature extraction but not understand the importance of how to organize these feature. We use two feature, HIS color information and CSS(Curvature Scale Space) as shape information. And then use neural network to adjust the weigh of the feature information. As a result, we show that good results can be obtained by two features and neural network. When we use that arithmetic, we could find the image that we need exactly and efficiently.
出处 《微计算机信息》 2010年第21期228-230,共3页 Control & Automation
基金 基金申请人:李峰 项目名称:基于多进制小波变换的虹膜识别方法研究 基金颁发部门:湖南省高等学校科学基金委(08A001) 基金申请人:李峰 项目名称:基于二维EMD的纹理分析方法研究 基金颁发部门:湖南省自然科学基金委(09JJ3120) 基金申请人:殷苌茗 项目名称:激励学习智能体最优控制策略及其在微经济环境下的决策问题 基金颁发部门:湖南省教育厅科学研究项目(07C083)
关键词 图像检索 特征提取 特征权重 神经网络 Image retrieval Feature extraction Feature weigh neural network
  • 相关文献

参考文献8

  • 1Veltkamp R C,Tanase M. Content-based image retrieval systems: a survey, UU-CS-2000-34 [D]. Department of Computer Science, Utretch University,October,2000. 被引量:1
  • 2Ritendra Datta, Jia Li.Content-based image retrieval approaches and trends of the new age [C].Proceedings of the 7th International Workshop on Multimedia Information Retrieval, in conjunction with ACM International Conference on Multimedia, Singapore, ACM, November 2005:253-262. 被引量:1
  • 3Wang Yi,Zhai Hong-chen. Shape description matrix and its applications to color-image retrieval and recognition [J].Science in China Ser.E Engineering & Materials Science, 2004,34(3):337-344. 被引量:1
  • 4Eauqueur J,Boujemaa N. Region-based image retrieval:fast coarse segmentation and fine color description[J]. Journal of Vision Languages and Computing.Special Issue on Vision Information System,200d, 15(1):69-95. 被引量:1
  • 5Ediz Saykol. A histogram-based approach for object-based query-by-shape and color in image and video databases[J]. Image and Vision Computing, 2005, 23:1170-1180. 被引量:1
  • 6Jeong S, Won C S. Image retrieval using color histograms generated by Gauss mixture vector quantization [J]. Computer Vision and Image Understanding, 2004, 9(1-3):44-46. 被引量:1
  • 7Jeong-Yo Ha, Gye-Young Kim. The Content-based Image Retrieval Method Using Multiple Features[C]. The 4th International Conference on Networked Computing and Advanced Information Management. 2008:652-657. 被引量:1
  • 8杨晓强.基于颜色量化与聚类的图像检索算法[J].微计算机信息,2008(12):286-287. 被引量:6

二级参考文献3

共引文献5

同被引文献10

引证文献2

二级引证文献3

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