计算机科学 ›› 2014, Vol. 41 ›› Issue (5): 288-291.doi: 10.11896/j.issn.1002-137X.2014.05.061

• 图形图像与模式识别 • 上一篇    下一篇

基于属性学习的图像分类研究

林武旭,成科扬,张建明   

  1. 江苏大学计算机科学与通信工程学院 镇江212013;江苏大学计算机科学与通信工程学院 镇江212013;江苏大学计算机科学与通信工程学院 镇江212013
  • 出版日期:2018-11-14 发布日期:2018-11-14
  • 基金资助:
    本文受国家自然科学基金(61170126)资助

Research on Image Classification Based on Attribute Learning

LIN Wu-xu,CHENG Ke-yang and ZHANG Jian-ming   

  • Online:2018-11-14 Published:2018-11-14

摘要: 图像中所蕴含的属性对于图像识别有着重要作用,以往的传统分类方法往往忽略了这些特性,为此,提出一种将稀疏表示和属性学习结合用于图像分类的新方法。该方法首先对图像特征进行稀疏分解,利用系数稀疏表示重构图像特征,然后将重构的特征数据用于属性学习,通过属性分类器的训练学习完成对目标图像的属性识别,达到识别出图像种类的目的。在植物数据集上的对比试验证实了该算法的有效性和在识别准确率上相对于传统识别算法的提升。

关键词: 属性学习,稀疏表示,多分类,K-SVD

Abstract: Inherent attributes of the image play an important role in image recognition,but the traditional classification methods tend to overlook these characteristics.This paper presented a method for image classification which combines the sparse representation with attribute learning.First the sparse representation codes of the image features are calcula-ted,then these codes are used for rebuilding the image features,finally these new features are used to train attribute classifiers which work for the purpose of image classification.Experiments using plant datasets show the efficiency of our method and improvement over the classical method.

Key words: Attribute learning,Sparse representation,Multiple classification,K-SVD

[1] Lampert C H,Nickisch H,Harmeling S.Learning To DetectUnseen Object Classes by Between-Class Attribute Transfer[C]∥CVPR.2009
[2] Lowe D G.Distinctive Image Features from Scale-invariant Keypoints[J].IJCV,2004,60(2):91-110
[3] Biederman I.Recognition by components-a theory of human ima-ge understanding[J].Psychological Review,1987,94(2):115-147
[4] Bosch A,Zisserman A,Munoz X.Representing shape with aspatial pyramid kernel[C]∥InCIVR.2007
[5] Farhadi A,Endres I,Hoiem D,et al.Describing Objects by their Attributes[C]∥CVPR.2009
[6] Rohrbach M,Stark M,Szarvas G,et al.What Helps Where And Why? Semantic Relatedness for Knowledge Transfer[C]∥CVPR.2010
[7] Berg T L,Berg A C,Shih J.Automatic Attribute Discovery and Characterization from Noisy Web Data[C]∥ECCV.2010
[8] Parikh D,Grauman K.Interactively Building a DiscriminativeVocabu-lary of Nameable Attributes[C]∥CVPR.2011
[9] Russakovsky O,Fei-Fei L.Attribute Learning in Large-scaleDatasets.Workshop on Parts and Attributes[C]∥ECCV.2010
[10] Yu Xiao-dong,Aloimonos Y.Attribute-based Transfer Learning for Object Categorization with Zero or One Training Example[C]∥ECCV.2010
[11] Aharon M,Elad M,Bruckstein A.K-SVD:An Algorithm for Designing Overcomplete Dictionaries for Sparse Representation[J].IEEE Transactions on Signal Processong,2006,54(11):4311-4322
[12] Rubinstein R,Zibulevsky M,Elad M.Efficient Implementation of the K-SVD Algorithm using Batch Orthogonal Matching Pursuit[R].Technion-Computer Science Department-technical Report CS-2008-08.revised-2008
[13] Nilsback M-E,Zisserman A.A Visual Vocabulary for FlowerClassification[C]∥Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2006
[14] Stark M,Goesele M,Schiele B.A Shape-based Object ClassModel for Knowledge Transfer[C]∥ICCV.2009
[15] Wang G,Forsyth D.Joint Learning of Visual Attributes Object Classes and Visual Salience[C]∥ICCV.2009
[16] 宋鑫颖,周志逵.一种基于SVM的主动学习文本分类方法[C]∥第二十三届中国数据库学术会议论文集(技术报告篇).2006

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!