电信科学 ›› 2016, Vol. 32 ›› Issue (4): 92-102.doi: 10.11959/j.issn.1000-0801.2016088

• 研究与开发 • 上一篇    下一篇

基于A-ELM的移动视觉搜索方法

胡海洋1,2,许军1,2,胡华1,2   

  1. 1 杭州电子科技大学计算机学院,浙江 杭州 310018
    2 杭州电子科技大学复杂系统建模与仿真教育部重点实验室,浙江 杭州 310018
  • 出版日期:2016-04-20 发布日期:2016-04-28
  • 基金资助:
    国家自然科学基金资助项目;国家自然科学基金资助项目;南京大学计算机软件新技术国家重点实验室开放基金资助项目;浙江省哲学社会科学重点研究基地(信息化与经济社会发展研究中心)课题资助项目

Mobile visual searching method based on ascending extreme learning machine

Haiyang HU1,2,Jun XU1,2,Hua HU1,2   

  1. 1 School of Computer Science and Technology,Hangzhou Dianzi University,Hangzhou 310018,China
    2 Key Laboratory of Complex Systems Modeling and Simulation,Ministry of Education,Hangzhou Dianzi University,Hangzhou 310018,China
  • Online:2016-04-20 Published:2016-04-28
  • Supported by:
    The National Natural Science Foundation of China;The National Natural Science Foundation of China;The Foundation of State Key Laboratory for Novel Software Technology of Nanjing University;The Foundation of Key Research Base for Philosophy and Social Sciences of Zhejiang Province(Research Center of Information Technology & Economic and Social Development)

摘要:

计算机智能技术在图像领域已经得到广泛的应用。极限学习机(ELM)作为一种新兴技术,克服了其他传统智能技术所面临的一些问题,吸引了越来越多研究人员的关注。首先对ELM算法的性能进行了分析验证,并将其延伸到图像分类搜索上。在此基础上,提出了基本视觉搜索(BMVS)框架,将ELM运用到此框架服务器端,并进一步优化了ELM的分类性能。最后实验证明ELM在移动视觉搜索方面的可行性,并通过和支持向量机(SVM)的实验对比验证相关方法的高效性。

关键词: 分类, 极限学习机, 移动视觉搜索

Abstract:

Computer intelligence technology had been widely used in the field of image searching. Extreme learning machine has emerged as a new technology which overcomes the problems in traditional intelligent field and it has attracted more and more researchers. The algorithm performance of ELM was analyzed firstly,extending the method to image classification field. A basic mobile visual searching(BMVS)framework was proposed which applies ELM to image searching and optimizes the performance of ELM. Finally,the experiment proves the effectiveness of the method proposed by using ELM for the mobile vision searching. Through the experiments of comparison with SVM-based methods,the efficiency of the method proposed was also confirmed.

Key words: classification, extreme learning machine, mobile visual searching

No Suggested Reading articles found!