基于上下文信息和核熵成分分析的目标分类算法

doi:10.3969/j.issn.0372-2112.2016.03.013

电子学报 ›› 2016, Vol. 44 ›› Issue (3): 580-586.DOI: 10.3969/j.issn.0372-2112.2016.03.013

基于上下文信息和核熵成分分析的目标分类算法

潘泓^1,2, 朱亚平³, 夏思宇^1,2, 金立左^1,2

1. 东南大学自动化学院, 江苏南京 210096;
2. 东南大学复杂工程系统测量与控制教育部重点实验室, 江苏南京 210096;
3. 中国传媒大学信息工程学院, 北京 100024

收稿日期:2014-07-18 修回日期:2014-11-21 出版日期:2016-03-25
- 通讯作者:
- 潘泓
- 作者简介:
- 朱亚平 女,1977年生于河南.现为中国传媒大学信息工程学院教授.主要研究方向为图像处理和机器学习. E-mail:zhuyaping@cuc.edu.cn;夏思宇 男,1978年生于湖北.现为东南大学自动化学院副教授.主要研究方向为机器学习,模式识别,社交媒体分析等. E-mail:xsy@seu.edu.cn;金立左 男,1972年生于江苏.现为东南大学自动化学院副教授.主要研究方向机器视觉、模式识别、机器学习、信息融合与嵌入式系统. E-mail:jinlizuo@gmail.com
- 基金资助:
- 国家自然科学基金 (No.61101165); 江苏省自然科学基金 (No.BK20131296,No.BK20130639); 航空科学基金 (No.20115169016); 国家科技支撑计划课题 (No.2012BAH02B03)

Object Classification Using Context Cue and Kernel Entropy Component Analysis

PAN Hong^1,2, ZHU Ya-ping³, XIA Si-yu^1,2, JIN Li-zuo^1,2

1. School of Automation, Southeast University, Nanjing, Jiangsu 210096, China;
2. Key Laboratory of Measurement and Control of CSE, Ministry of Education, Southeast University, Nanjing, Jiangsu 210096, China;
3. School of Information Engineering, Communication University of China, Beijing 100024, China

Received:2014-07-18 Revised:2014-11-21 Online:2016-03-25 Published:2016-03-25

摘要/Abstract

摘要：

结合图像属性上下文信息和核熵成分分析,构造了一种新颖的基于下上文信息的局部特征描述子——上下文核描述子(Context Kernel Descriptors,CKD).上下文信息的引入提高了CKD特征的鲁棒性,减少了特征误匹配.核熵成分分析从全维CKD特征分量中选出最能代表目标几何结构信息的特征分量,将其投影到这些特征分量张成的子空间上可得到降维CKD特征.在Caltech-101和CIFAR-10的测试结果表明,CKD的分类性能不仅明显优于其它局部特征描述子,还优于多数基于稀疏表示和深度学习等复杂模型的目标分类算法.

关键词: 上下文核描述子, 核熵成分分析, 特征降维, 目标分类

Abstract:

Combining the context cue of image attributes and kernel entropy component analysis (KECA),we proposed a context-based local feature descriptor called context kernel descriptors (CKD).Context cue implied in the CKD improves its robustness,thus reducing false matches during feature correspondence.KECA applied in the feature dimensionality reduction step selects the principal eigenvectors that contribute most to the geometrical structure of input images.Projecting the full-dimensional CKD onto the subspaces spanned by these principal eigenvectors,we derive the final low-dimensional CKD.Evaluation results on Caltech-101 and CIFAR-10 show that the classification performance of the proposed CKD significantly outperforms other local descriptors,and even surpasses most sparse representation-based and deep learning-based sophisticated object classification methods.

Key words: context kernel descriptors, kernel entropy component analysis, feature reduction, object classification

中图分类号:

TP391

潘泓, 朱亚平, 夏思宇, 金立左. 基于上下文信息和核熵成分分析的目标分类算法[J]. 电子学报, 2016, 44(3): 580-586.

PAN Hong, ZHU Ya-ping, XIA Si-yu, JIN Li-zuo. Object Classification Using Context Cue and Kernel Entropy Component Analysis[J]. Acta Electronica Sinica, 2016, 44(3): 580-586.

参考文献

[1] Lowe D.Distinctive image features from scale-invariant keypoints[J]. International Journal of Computer Vision,2004,60(2):91-110.
[2] Dalal N,Triggs B.Histograms of oriented gradients for human detection[A]. Schmid C.Proceedings of CVPR[C]. San Diego,CA,USA:IEEE,2005.886-893.
[3] Alcantarilla P,Bartoli A,Davison A.KAZE features[A]. Fitzgibbon A.Proceedings of ECCV[C]. Florence,Italy:Springer,2012.214-227.
[4] Ojala T,Pietikäinen M,Mäenpää T.Multiresolution gray-scale and rotation invariant texture classification with local binary patterns[J]. IEEE Trans,2002,PAMI-24(7):971-987.
[5] Kannala J,Rahtu E.BSIF:Binarized statistical image features[A]. Proceedings of ICPR[C]. Tsukuba,Japan:IEEE,2012.1363-1366.
[6] Pedersen K,Smidt K,Ziem A,et al.Shape index descriptors applied to texture-based galaxy analysis[A]. Proceedings of ICCV[C] ,Sydney,NSW,Australia:IEEE,2013.2240-2447.
[7] Belongie S,Malik J,Puzicha J.Shape matching and object recognition using shape contexts[J]. IEEE Trans,2002,PAMI-24(4):509-522.
[8] Zitnick C.Binary coherent edge descriptors[A]. Daniilidis K.Proceedings of ECCV[C]. Heraklion,Crete,Greece:Springer,2010.170-182.
[9] Scholkopf B,Smola A,Mulle K.Kernel principal component analysis[A]. Gerstner W.Proceedings of ICANN[C]. Lausanne,Switzerland:Springer,1997.583-588.
[10] Bo L,Sminchisescu C.Efficient match kernel between sets of features for visual recognition[A]. Bengio Y.Proceedings of NIPS[C]. Vancouver,Canada:NIPS,2009.135-143.
[11] Lazebnik S,Schmid C,Ponce J.Beyond bags of features:spatial pyramid matching for recognizing natural scene categories[A]. Fitzgibbon A.Proceedings of CVPR[C]. New York,NY,USA:IEEE,2006.2169-2178.
[12] Grauman K,Darrell T.The pyramid match kernel:Efficient learning with sets of features[J]. JMLR,2007,8(4):725-760.
[13] Galleguillos C,Belongie S.Context based object categorization:A critical survey[J]. Computer Vision and Image Understanding,2010,114(6):712-722.
[14] Myung J,Torralba A,Willsky A.Context models and out-of-context objects[J]. Pattern Recognition Letters,2012,33(7):853-862.
[15] Bo L,Ren X,Fox D.Kernel descriptors for visual recognition[A]. Lafferty J.Proceedings of NIPS[C]. Vancouver,Canada:NIPS,2010.244-252.
[16] Wang P,Wang J,Zeng G,et al.Supervised kernel descriptors for visual recognition[A]. Proceedings of CVPR[C]. Portland,Oregon,USA:IEEE,2013.2858-2865.
[17] Jenssen R.Kernel entropy component analysis[J]. IEEE Trans,2010,PAMI-32(5):847-860.
[18] Principe J.Information Theoretic Learning:Renyi's Entropy and Kernel Perspectives[M]. New York:Springer,2010.
[19] Rényi A.On measures of entropy and information[A]. Neyman J.Fourth Berkeley Symposium on Mathematical Statistics and Probability[C]. Berkeley,CA,USA:University of California Press,1961.547-561.
[20] Parzen E.On the estimation of a probability density function and the mode[J]. The Annals of Mathematical Statistics,1962,33(3):1065-1076.
[21] Li F,Fergus R,Perona P.One-shot learning of object categories[J]. IEEE Trans,2006,PAMI-28(4):594-611.
[22] Jia Y,Huang C,Darrell T.Beyond spatial pyramids:Receptive field learning for pooled image features[A]. Proceedings of CVPR[C]. Providence,Rhode Island,USA:IEEE,2012.3370-3377.
[23] McCann S Lowe D.Spatially local coding for object recognition[A]. Lee M K.Proceedings of ACCV[C]. Daejeon,Korea:Springer,2012.204-217.
[24] Zeiler M,Taylor W,Fergus R.Adaptive deconvolutional networks for mid and high level feature learning[A]. Schiele B.Proceedings of ICCV[C]. Barcelona,Spain:IEEE,2011.2018-2025.
[25] Boureau L,Roux L,Bach F,et al.Ask the locals:Multi-way local pooling for image recognition[A]. Schiele B.Proceedings of ICCV[C]. Barcelona,Spain:IEEE,2011.2651-2658.
[26] Liu L,Wang L,Liu X.In defense of soft-assignment coding[A]. Schiele B.Proceedings of ICCV[C]. Barcelona,Spain:IEEE,2011.2486-2493.
[27] Feng J,Ni B,Tian Q,et al.Geometriclp-norm feature pooling for image classification[A]. Proceedings of CVPR[C]. Colorado Springs,USA:IEEE,2011.2697-2704.
[28] Jiang Z,Zhang G,Davis S.Submodular dictionary learning for sparse coding[A]. Proceedings of CVPR[C]. Providence,Rhode Island,USA:IEEE,2012.3418-3425.
[29] Oliveira G,Nascimento E,Vieira A.Sparse spatial coding:a novel approach for efficient and accurate object recognition[A]. Proceedings of ICRA[C]. St Paul,MN,USA:IEEE,2012.2592-2598.
[30] Bo L,Ren X,Fox D.Multipath sparse coding using hierarchical matching pursuit[A]. Proceedings of CVPR[C]. Portland,Oregon,USA:IEEE,2013.660-667.
[31] Torralba A,Fergus R,Freeman W.80 million tiny images:A large data set for nonparametric object and scene recognition[J]. IEEE Trans,2008,PAMI-30(11):1958-1970.
[32] Yu K,Zhang T.Improved local coordinate coding using local tangents[A]. Proceedings of ICML[C]. Haifa,Israel:Omnipress,2010.1215-1222.
[33] Goodfellow I,Courville A,Bengio Y.Spike-and-Slab sparse coding for unsupervised feature discovery[A]. NIPS Workshop on Challenges in Learning Hierarchical Models[C]. Granada,Spain:NIPS,2011.
[34] Ciresan D,Meier U Schmidhuber J.Multi-columndeep neural networks for image classification[A]. Proceedings of CVPR[C]. Providence,Rhode Island,USA:IEEE,2012.3642-3649.
[35] Zeiler M,Fergus R.Stochastic pooling for regularization of deep convolutional neural networks[A]. Goodfellow I.Proceedings of ICLR[C]. Scottsdale,Arizona,USA:IEEE,2013.
[36] Le Q,Ngiam J,Chia C,et al.Tiled convolutional neural networks[A]. Lafferty J.Proceedings of NIPS[C]. Vancouver,Canada:NIPS,2010.1279-1287.
[37] Gens R,Domingos P.Discriminative learning of sum-product networks[A]. Pereira F.Proceedings of NIPS[C]. Lake Tahoe,Nevada,USA:NIPS,2012.1-9.
[38] Bo L,Lai K,Ren X,et al.Object recognition with hierarchical kernel descriptors[A]. Proceedings of CVPR[C]. Colorado Springs,USA:IEEE,2011.1729-1736.e S. Context based object categorization: A critical survey [J]. Computer Vision and Image Understanding, 2010, 114:712-722.
[46]. Torralba A, Murphy K, Freeman, et al. Context-Based Vision System for Place and Object Recognition [C]. Proceedings of ICCV, 2003, 1:273-280.
[47]. Myung J, Torralba A, and Willsky A. A tree-based context model for object recognition [J]. IEEE Trans. Pattern Analysis and Machine Intelligence, 2012, 34(2):240–252.
[48]. Myung J, Torralba A, and Willsky A. Context models and out-of-context objects [J]. Pattern Recognition Letters, 2012, 33(7):853-862.

基于上下文信息和核熵成分分析的目标分类算法

Object Classification Using Context Cue and Kernel Entropy Component Analysis

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 3

编辑推荐

Metrics

[1]	赖裕平, 高宁, 何闻达, 平原, 杜春来, 王宝成, 丁洪伟. 贝塔混合模型的变分贝叶斯学习及应用[J]. 电子学报, 2018, 46(7): 1787-1792.
[2]	高妮, 高岭, 贺毅岳, 王海. 基于自编码网络特征降维的轻量级入侵检测模型[J]. 电子学报, 2017, 45(3): 730-739.
[3]	赵永威, 周苑, 李弼程, 柯圣财. 基于近义词自适应软分配和卡方模型的图像目标分类方法[J]. 电子学报, 2016, 44(9): 2181-2188.