Please wait a minute...
浙江大学学报(工学版)
浙江大学学报(工学版)  2017, Vol. 51 Issue (12): 2311-2319    DOI: 10.3785/j.issn.1008-973X.2017.12.002
计算机与通信技术     
BOW-HOG特征图像分类
邹北骥1, 郭建京1, 朱承璋2, 杨文君2, 吴慧2, 何骐2
1. 中南大学 信息科学与工程学院, 湖南 长沙 410083;
2. 中南大学 眼科医学影像处理研究中心, 湖南 长沙 410083
Image classification based on BOW-HOG feature
ZOU Bei-ji1, GUO Jian-jing1, ZHU Cheng-zhang2, YANG Wen-Jun2, WU Hui2, HE Qi2
1. School of Information Science and Engineering, Central South University, Changsha 410083, China;
2. Center for Ophthalmic Imaging Research, Central South University, Changsha 410083, China
 全文: PDF(4985 KB)   HTML
摘要：

为了减少图像尺寸对提取特征的影响，同时移除特征向量中的冗余信息，将词汇袋模型（BOW）与梯度方向直方图（HOG）特征相结合，提出一种基于BOW-HOG的特征描述子用于图像分类.将图像划分为不同的子区域，对梯度幅值较大的子区域提取HOG特征.用BOW模型对子区域HOG特征编码，构建原始图像上维度一致的特征向量.将特征向量输入训练好的分类器，完成图像分类任务.将BOW-HOG特征描述子在不同的图像分类任务上进行试验，包括图像文本分类、图像场景分类.本实验的文本分类正确率为0.813，场景分类正确率为0.826，优于传统基于HOG特征的方法，表明了基于BOW-HOG特征图像分类方法的可行性、有效性.
Abstract:

A new feature descriptor named as BOW-HOG combining Bag of Words (BOW) model with HOG was proposed in order to reduce the influence of image size on feature extraction results,and remove the redundant information of feature vectors at the same time. Firstly, the input image was divided into series of sub-regions, and HOG features were extracted from sub-regions with large gradient values. Then, the BOW model was applied to encode the obtained HOG features; the whole feature vector of the original image was constructed. Finally, the feature vector was put into a trained classifier for image texts or image scenes classification. The BOW-HOG feature descriptor was applied on some image classification tasks, including image text classification and image scene classification. The experimental results show that the proposed method achieves a text classification accuracy of 0.813 and a scene classification accuracy of 0.826,respectively,which outperforms the traditional HOG based methods, indicating the feasibility and effectiveness of the proposed BOW-HOG based image classification method.
收稿日期: 2016-10-25 出版日期: 2017-11-22
CLC:  TP394.1  
基金资助:

国家自然科学基金资助项目（61573380，61262032）.
通讯作者: 朱承璋,女,讲师,博士.orcid.org/0000-0002-6652-8903.     E-mail: anandawork@126.com
作者简介: 邹北骥(1961-),男,教授,从事计算机视觉研究.orcid.org/0000-0002-3542-1097.E-mail:bjzou@csu.edu.cn
服务　　
把本文推荐给朋友
加入引用管理器
E-mail Alert
作者相关文章　　

引用本文:

邹北骥, 郭建京, 朱承璋, 杨文君, 吴慧, 何骐. BOW-HOG特征图像分类[J]. 浙江大学学报(工学版), 2017, 51(12): 2311-2319.

ZOU Bei-ji, GUO Jian-jing, ZHU Cheng-zhang, YANG Wen-Jun, WU Hui, HE Qi. Image classification based on BOW-HOG feature. JOURNAL OF ZHEJIANG UNIVERSITY (ENGINEERING SCIENCE), 2017, 51(12): 2311-2319.

链接本文:

http://www.zjujournals.com/eng/CN/10.3785/j.issn.1008-973X.2017.12.002    或    http://www.zjujournals.com/eng/CN/Y2017/V51/I12/2311

[1] 葛娟,曹伟国,周炜,等.一种颜色仿射变换下的局部特征描述子[J].计算机辅助设计与图形学学报,2013,25(1):26-33. GE Juan, CAO Wei-guo, ZHOU Wei, et al. A local feature descriptor under color affine transformation[J]. Journal of Computer-Aided Design and Computer Graphics, 2013, 25(1):26-33.
[2] LIENHART R, MAYDT J. An extended set of Haar-like features for rapid object detection[C]//International Conference on Image Processing 2002. New York:ICIP, 2002:I-900-I-903.
[3] DO T T, DOAN K N, LE T H, et al. Boosted of Haar-like features and local binary pattern based face detection[C]//International Conference on Computing and Communication Technologies. Danang:IEEE, 2009:1-8.
[4] MITA T, KANEKO T, HORI O. Joint Haar-like features for face detection[C]//IEEE International Conference on Computer Vision. Beijing:ICCV, 2005:1619-1626.
[5] OJALA T, HARWOOD I. A comparative study of texture measures with classification based on feature distributions[J]. Pattern Recognition, 1996, 29(1):51-59.
[6] OJALA T, PIETIK, INEN M, et al. Multiresolution gray-scale and rotation invariant texture classification with local binary patterns[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002,24(7):971-987.
[7] LOWE D G. Distinctive image features from scale-invariant keypoints[J]. International Journal of ComputerVision, 2004, 60(2):91-110.
[8] DALAL N, TRIGGS B. Histograms of oriented gradients for human detection[C]//IEEE Computer Society Conference on Computer Vision and Pattern Recognition. San Diego:IEEE, 2005:886-893.
[9] WANG X, HAN T X, YAN S. An HOG-LBP human detector with partial occlusion handling[C]//IEEE International Conference on Computer Vision. Kyoto:IEEE, 2009:32-39.
[10] KARAABA M F, SURINTA O, SCHOMAKER L R B, et al. Robust face identification with small sample sizes using bag of words and histogram of oriented gradients[C]//International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications. Beijing:2016:582-589.
[11] MINETTO R, THOME N, CORD M, et al. T-HOG:an effective gradient-based descriptor for single line text regions[J]. Pattern Recognition, 2013, 46(3):1078-1090.
[12] BANERJI S, SINHA A, LIU C. HaarHOG:improving the HOG descriptor for image classification[C]//IEEE International Conference on Systems, Man, and Cybernetics. San Diego:IEEE, 2013:4276-4281.
[13] WU H, ZOU B, ZHAO Y Q, et al. Scene text detection using adaptive color reduction, adjacent character model and hybrid verification strategy[J]. Visual Computer International Journal of Computer Graphics, 2017, 33(1):113-126.
[14] SINHA A, BANERJI S, LIU C. New color GPHOGdescriptors for object and scene image classification[J]. Machine Vision and Applications, 2014, 25(2):361-375.
[15] SUYKENS J A K, VANDEWALLE J. Least squares support vector machine classifiers[J]. Neural Processing Letters, 1999, 9(3):293-300.
[16] SCHAPIRE R E, SINGER Y. Improved boostingalgorithms using confidence-rated predictions[J]. Machine Learning, 1999, 37(3):297-336.
[17] CUTLER A, CUTLER D R, STEVENS J R. Random forests[J]. Machine Learning, 2011, 45(1):157-176.
[18] MATAS J, CHUM O, URBAN M, et al. Robust wide-baseline stereo from maximally stable extremal regions[J]. Image and Vision Computing, 2004,22(10):761-767.
[19] SUNG M C, JUN B, CHO H, et al. Scene text detection with robust character candidate extraction method[C]//201513th International Conference on Document Analysis and Recognition (ICDAR). Nancy:ICDAR,2015:426-430.
[20] LI L J, LI F F. What, where and who classifyingevents by scene and object recognition[C]//IEEEInternational Conference on Computer Vision. Rio de Janeiro:IEEE. 2007:1-8.
[21] ZHANG Z, SHEN W, YAO C, et al. Symmetry-based text line detection in natural scenes[C]//IEEE Conference on Computer Vision and Pattern Recognition. Boston:IEEE, 2015:2558-2567.
[22] GAO X, HUA G, NIU Z, et al. Context aware topic model for scene recognition[C]//IEEE Conference on Computer Vision and Pattern Recognition. Portland:IEEE, 2012:2743-2750.
[23] OLIVA A, TORRALBA A. Modeling the shape of the scene:a holistic representation of the spatial envelope[J]. International Journal of Computer Vision, 2001, 42(3):145-175.
[24] SU H, XING E P, LI F F, et al. Object bank:a high-level image representation for scene classification andsemantic feature sparsification[J]. Advances in NeuralInformation Processing Systems, 2010, 26:1378-1386.
[25] VAN GEMERT J C, VEENMAN C J, SMEULDERS A W, et al. Visual word ambiguity[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 32(7):1271-1283.
No related articles found!