利用编码层特征组合进行场景分类

吉林大学学报(工学版) ›› 2013, Vol. 43 ›› Issue (增刊1): 450-454.

利用编码层特征组合进行场景分类

崔崟¹, 段菲², 章毓晋²

1. 北京航空航天大学电子信息工程学院,北京100191;
2. 清华大学电子工程系,北京100084

收稿日期:2012-08-27 发布日期:2013-06-01
作者简介:崔崟(1990-),男,博士研究生.研究方向:图像分类.E-mail:richardaecn@gmail.com
基金资助:
国家自然科学基金项目(61171118);教育部高等学校博士学科点专项科研基金项目(SRFDP-20110002110057).

Scene classification based on coding layer feature combination

CUI Yin¹, DUAN Fei², ZHANG Yu-jin²

1. School of Electronic Information Engineering, Beihang University, Beijing 100191, China;
2. Department of Electronic Engineering, Tsinghua University, Beijing 100084, China

Received:2012-08-27 Published:2013-06-01

摘要/Abstract

摘要：

为了提高图像场景分类效果,一种有效的方法是将不同特征组合起来,以常用的SIFT特征和归一化颜色直方图(NCH)特征为例研究了不同语义层次上的组合——特征层的组合和编码层的组合以及它们的效果。在几个常用数据库上的实验结果表明,相比简单的组合特征,在特征提取后再进行组合能在降低特征维数的情况下保持分类的效果,而在编码层的特征组合能获得更高的分类准确率。这表明特征组合应尽量在较高的语义层次上进行。

关键词: 场景分类, 词袋模型, 特征组合, 语义层次

Abstract:

In order to improve classification performance of image,one efficient method was to combine different features together.The combination in various layers (feature layer and coding layer) was studied with the help of SIFT feature and normalized color histogram (NCH) feature,their effects were also compared.Experiments on several popular datasets show that,compared to combined features,the feature combination after feature extraction can reduce feature dimension while keep the similar classification rate,and the feature combination on coding layer can achieve even higher classification rate.It is concluded that feature combination should be better performed on higher semantic level.

Key words: scene classification, bag of words, feature combination, semantic level

中图分类号:

TN911.73

崔崟, 段菲, 章毓晋. 利用编码层特征组合进行场景分类[J]. 吉林大学学报(工学版), 2013, 43(增刊1): 450-454.

CUI Yin, DUAN Fei, ZHANG Yu-jin. Scene classification based on coding layer feature combination[J]. 吉林大学学报(工学版), 2013, 43(增刊1): 450-454.

参考文献

[1] 章毓晋.图像工程(下册):图像理解[M].(第3版).北京:清华大学出版社,2012.

[2] Oliva A,Torralba A.Modeling the shape of the scene:a holistic representation of the spatial envelope [J].IJCV,2001,42(3):145-175.

[3] Sivic J,Zisserman A.Video Google:A text retrieval approach to object matching in videos [C]//Proc ICCV II,Nice,France,2003:1470-1477.

[4] Li F F,Perona P.A bayesian hierarchical model for learning natural scene categories [C]//Proc CVPR,San Diego,USA,2005:524-531.

[5] Bosch A,Zisserman A,Munoz X.Scene classi cation via pLSA [C]//Proc ECCV,Graz, Austria,2006:517-530.

[6] Blei D,Ng A,Jordan M.Latent dirichlet allocation [J].Journal of Machine Learning Research,2003(3):993-1022.

[7] Sivic J,Russell B C,Efros A A,et al.Discovering objects and their location in images [C]//Proc ICCV,Beijing,China,2005:370-377.

[8] Lazebnik S,Schmid C,Ponce J.Beyond bags of features:Spatial pyramid matching for recognizing natural scene categories [C]//Proc CVPR,New York,2006:2169-2178.

[9] Yang J,Yu K,Gong Y,et al.Linear spatial pyramid matching using sparse coding for image classification [C]//Proc CVPR,2009:1794-1801.

[10] Lowe D.Distinctive image features from scale-invariant key points [J].IJCV,2004,60(2):91-110.

[11] Nixon M S,Aguado A S.Feature extraction and image processing [M].(2ed).Academic Press,2008.

[12] Van de Sande K E A,Gevers T,Snoek C G M.Evaluation of color descriptors for objects and scene recognition [C]// Proc CVPR,Anchorage,USA,2008:1-8.

[13] Van Gemert J C,Veenman C J,Smeulders A W M,et al.Visual word ambiguity [J].TPAMI,2010,32(7):1271-1283.

[14] Boureau Y,Bach F,Le Cun Y,et al.Learning mid-level features for recognition [C]//Proc CVPR,2010:2559-2566.

[15] Maji S,Berg A C,Malik J.Classification using intersection kernel support vector machine is efficient [C]//Proc CVPR,Anchorage,USA,2008:1-8.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed