欢迎访问昆明冶金高等专科学校学报官方网站,今天是 分享到:

昆明冶金高等专科学校学报 ›› 2018, Vol. 34 ›› Issue (3): 59-65.DOI: 10.3969/j.issn.1009-0479.2018.03.01

• 电子信息技术 • 上一篇    下一篇

面向人物图像的近似检索算法性能评测

罗玉梅,王 瑞   

  1. 昆明冶金高等专科学校计算机信息学院,云南 昆明 650033
  • 收稿日期:2018-05-10 出版日期:2018-06-30 发布日期:2018-06-30
  • 作者简介:罗玉梅(1982-),女,云南楚雄人,讲师,软件工程硕士,主要从事计算机软件技术方向的教学研究。

Performance Evaluation of Near-duplicate Web Image Retrieval Algorithm of Person Images

LUO Yumei,WANG Rui   

  1. Faculty of Compute Information,Kunming Metallurgy College,Kunming 650033,China
  • Received:2018-05-10 Online:2018-06-30 Published:2018-06-30

摘要: 最新研究表明,互联网上存在的海量近似重复图像可以为解决一些传统上很困难的计算机视觉任务提供新的解决方案。介绍了4个主流的用来进行近似重复图像检索的方法:Hash码,MeanSSIM,SIFT视觉词袋模型(BoVW)和属性关系图(ARG)。构建了一个包含24762幅图像的与人物有关的图像数据集,观察数据确定了4种近似重复图像类型。利用该数据集通过实验定量评估了4种近似重复图像检索方法的运行效率和检索精度,最后推荐使用基于SIFT特征的视觉词袋方法来进行面向人物图像的近似重复Web图像检索。

关键词: 检索, Web图像, 算法, 评测

Abstract: Recent studies have revealed that numerous near-duplicate Web images can be used as an intermediate step to implement some traditional difficult compute rvision tasks.This paper presents a comprehensive study of the existing near-duplicate image retrieval methods in a structural way.Four representatives of the existing methods,i.e.hash signature,meanSSIM,BoVW  with SIFT features and ARG,are experimentally evaluated using a self-constructed data set containing 24762 images.The experimental results reveal that compared with global feature based methods,local feature based ones are usually more appropriate for the task of person identification in web images,as they can deal with partial duplicate and scene similar images better.In particular,BoVW  with SIFT features is recommended as it provides the best trade-offbetween on-line speed and retrieval accuracy.

Key words: retrieval, Web images, algorithm, evaluation

中图分类号: