期刊文献+

类别约束下的低秩优化特征字典构造方法 被引量:1

Low-rank optimization characteristic dictionary training approach with category constraint
下载PDF
导出
摘要 字典模型(BOW)是一种经典的图像描述方法,模型中特征字典的构造方法至关重要。针对特征字典构造问题,提出了一种类别约束下的低秩优化特征字典构造方法 LRC-DT,通过低秩优化的方法使训练出来的特征字典在描述同类图像时表示系数矩阵的秩相对较低,从而将类别信息引入到字典学习中,提高字典对图像描述的可分辨性。在标准公测库Caltech-101和Caltech-256上的实验结果表明:将SPM、稀疏编码下的SPM(ScSPM)、局部线性编码(LLC)和线性核函数的SPM(LSPM)编码方法中的特征字典替换为加入低秩约束(LRC)的特征字典后,随着训练样本数目增多,字典模型的分类准确率与未引入低秩约束的方法相比有所提高。 Bag Of Words( BOW) is a classical approach of image description, and the method of constructing the characteristic dictionary in this model is very important. A category constrained low-rank optimization characteristic dictionary training approach named LRC-DT was proposed for the characteristic dictionary construction. Through the low-rank optimization, the rank of the coefficient matrix constructed by same category images was minimized. Then the classification information was introduced into the characteristic dictionary learning to improve the identifiability of characteristic dictionary for image description. Some experiments were conducted on two standard image databases including Caltech-101 and Caltech-256,and the characteristic dictionary of SPM( Spatial Pyramid Matching), ScSPM( Sparse codes SPM), LLC( Locality-constrained Linear Coding) and LSPM( Linear SPM) were replaced by constrained low-rank optimization characteristic dictionary. The experimental results show that the proposed method can consistently offer better performance than not employing the category constrained low-rank optimization, its classification accuracy is improved with the increase of the training sample number.
出处 《计算机应用》 CSCD 北大核心 2014年第9期2668-2672,2677,共6页 journal of Computer Applications
基金 国土资源部公益性项目(201311006)
关键词 字典模型 低秩优化 低秩描述 图像描述 图像分类 Bag Of Words(BOW) low-rank optimization low-rank representation image representation image classification
  • 相关文献

参考文献8

  • 1Yuanyuan Liu,L.C. Jiao,Fanhua Shang,Fei Yin,F. Liu.An efficient matrix bi-factorization alternative optimization method for low-rank matrix recovery and completion[J].Neural Networks.2013 被引量:1
  • 2Jamie Shotton,John Winn,Carsten Rother,Antonio Criminisi.TextonBoost for Image Understanding: Multi-Class Object Recognition and Segmentation by Jointly Modeling Texture, Layout, and Context[J].International Journal of Computer Vision.2009(1) 被引量:1
  • 3Zhengdong Zhang,Arvind Ganesh,Xiao Liang,Yi Ma.TILT: Transform Invariant Low-Rank Textures[J].International Journal of Computer Vision.2012(1) 被引量:1
  • 4G.J. Burghouts,K. Schutte.Spatio-temporal layout of human actions for improved bag-of-words action detection[J].Pattern Recognition Letters.2013 被引量:1
  • 5Nan Zhang,Jian Yang.Low-rank representation based discriminative projection for robust feature extraction[J].Neurocomputing.2013 被引量:1
  • 6J. Zhang,M. Marsza?ek,S. Lazebnik,C. Schmid.Local Features and Kernels for Classification of Texture and Object Categories: A Comprehensive Study[J].International Journal of Computer Vision.2007(2) 被引量:1
  • 7邓承志,曹汉强.多尺度脊波字典的构造及其在图像编码中的应用[J].中国图象图形学报,2009,14(7):1273-1278. 被引量:6
  • 8孙玉宝,韦志辉,肖亮,张峥嵘,吕战强.多形态稀疏性正则化的图像超分辨率算法[J].电子学报,2010,38(12):2898-2903. 被引量:25

二级参考文献29

  • 1司菁菁,王成儒,程银波.基于改进的正交FRIT的分层图像编码算法[J].仪器仪表学报,2006,27(10):1283-1287. 被引量:1
  • 2舒正华,邹道文,汪胜前,刘国栋,邓松.一种基于ridgelet和wavelet的混合编码压缩算法[J].江西师范大学学报(自然科学版),2007,31(2):111-114. 被引量:1
  • 3Candes E J,Donoho D L.Curvelets-a surprsingly effective nonadaptive representation for objects with edges[A].In:Cohen A,Rabut C,Schumaker L,Editors:Curves and Surface Fitting[C],Nashville,Tennessee,USA:Vanderbilt University Press,1999. 被引量:1
  • 4Candes E J.Ridgelet:Theory and Applications[D].Department of Statistics,Stanford University,Stanford,CA,USA,1998. 被引量:1
  • 5Candes E J,Demanet L,Donoho D L,et al.Fast discrete curvelet transforms[J].Multiscale Modeling Simulation,2005,5 (3):861-899. 被引量:1
  • 6Do M N,Vetterli M.The contourlet transform:an efficient directional multiresolutian image representation[J].IEEE Transactions on Image Processing,2005,14(12):2091-2106. 被引量:1
  • 7Candes E J,Donoho D L.Ridgolets:A key to higher-dimensional intermittency[J].Philosophical Transactions of the Royal Society of London Series A,1999,357(1760):2495-2509. 被引量:1
  • 8Candes E J.Harmonic analysis of neural networks[J].Applied and Computational Harmonic Analysis,1999,6(2):197-218. 被引量:1
  • 9Candes E J.Monoscale ridgelet for the representation of images with edges[R].Technology Report,1999-27,Deptartment Statistic,Stanford University,Stanford,CA,USA,1999. 被引量:1
  • 10Mallat S,Zhang Z.Matching pursuit with time-frequency dictionaries[J].IEEE Transactions on Signal Processing,1993,41 (12):3397-3415. 被引量:1

共引文献29

同被引文献2

引证文献1

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