一种改进的二进制哈希编码方法

doi:10.3969/j.issn.0372-2112.2019.02.029

电子学报 ›› 2019, Vol. 47 ›› Issue (2): 462-469.DOI: 10.3969/j.issn.0372-2112.2019.02.029

一种改进的二进制哈希编码方法

江泽涛^1,2, 简雄^1,2, 刘小艳^1,2, 曾聪文^1,2

1. 桂林电子科技大学广西高校图像图形智能处理重点实验室, 广西桂林 541004;
2. 桂林电子科技大学广西可信软件重点实验室, 广西桂林 541004

收稿日期:2017-05-02 修回日期:2018-07-20 出版日期:2019-02-25
- 通讯作者:
- 简雄
- 作者简介:
- 江泽涛 男,1961年生于江西九江,教授,博士,研究方向为计算机视觉与信息安全.E-mail:zetaojiang@126.com
- 基金资助:
- 国家自然科学基金 (No.61572147，No.61762066）; 广西科技计划项目 (No.AC16380108）; 桂林电子科技大学图像图形智能处理重点实验室项目 (No.GⅡP201405，No.GⅡP201701）; 广西可信软件重点实验室项目 (No.kx201502）; 广西研究生教育创新计划 (No.2018YJCX46）; 桂林电子科技大学校级教改重点项目 (No.JGA201506）; 广西云计算与大数据协同创新中心 (No.YD16304）

An Improved Binary Hash Coding Method

JIANG Ze-tao^1,2, JIAN Xiong^1,2, LIU Xiao-yan^1,2, ZENG Cong-wen^1,2

1. Key Laboratory of Image and Graphic Intelligent Processing of Guangxi, Guilin University of Electronic Technology, Guilin, Guangxi 541004, China;
2. Key Laboratory of Dependable Software of Guangxi, Guilin University of Electronic Technology, Guilin, Guangxi 541004, China

Received:2017-05-02 Revised:2018-07-20 Online:2019-02-25 Published:2019-02-25

摘要/Abstract

摘要： 为了应对手工视觉特征与哈希编码过程不能最佳地兼容以及现有哈希方法无法区分图像语义信息的问题，提出一种基于深度卷积神经网络学习二进制哈希编码的方法.该方法基本思想是在深度残差网络中增加一个哈希层，同时学习图像特征和哈希函数；以此同时提出一种更加紧凑的分级哈希结构，用来提取更加接近图像语义的特征.经MNIST、CIFAR-10、NUS-WIDE数据集的实验，结果表明该方法优于现有的哈希方法.该方法不仅统一了特征学习和哈希编码的过程，同时深层残差网络也能得到更接近图像语义的特征，进而提高了检索准确度.

关键词: 图像检索, 深度残差网络, 分级哈希, 语义信息

Abstract: To address the problems that hand-engineering visual features can't be optimally compatible with the Hash coding process and existing Hash methods can't differentiate images semantics information,a learning method of binary Hashing based on deep convolutional neural networks is proposed.The basic idea is to add a Hash layer into the deep residual network and to learn simultaneously image features and Hash functions.Meanwhile,we propose a more compact hierarchical Hashing structure to extract features closer to semantics information of images.Experimental results of MNIST,CIFAR-10 and NUS-WIDE datasets show that the method is superior to existing Hashing methods.This method not only unifies the process of feature learning and Hash coding,and at the same time,the deep residual network is able to get features closer to image semantics.Thus the retrieval accuracy is improved.

Key words: image retrieval, deep residual network, hierarchical Hashing, semantics information

中图分类号:

TP391.41

江泽涛, 简雄, 刘小艳, 曾聪文. 一种改进的二进制哈希编码方法[J]. 电子学报, 2019, 47(2): 462-469.

JIANG Ze-tao, JIAN Xiong, LIU Xiao-yan, ZENG Cong-wen. An Improved Binary Hash Coding Method[J]. Acta Electronica Sinica, 2019, 47(2): 462-469.

参考文献

[1] Indyk P,Motwani R.Approximate nearest neighbors:towards removing the curse of dimensionality[A].Proceedings of the ACM Symposium on Theory of Computing[C].TX,USA:ACM,1998.604-613.
[2] Bentley JL.K-d trees for semi-dynamic pointsets[A].Proceedings of the Sixth Annual Symposium on Computational Geometry[C].Berkeley,Califor-nia,USA:ACM,1990.187-197.
[3] Silpaanan C,Hartley R.Optimised KD-trees forfast image descriptor matching[A].Proceedings ofthe IEEE Conference on Computer Vision and Pattern Recognition[C].Anchorage,Alaska,USA:IEEE,2008.1-8.
[4] Gionis A,Indyk P,Motwani R.Similarity searchin high dimensions via hashing[A].Proceedings of 25th International Conference on Very Large DataBases[C].Edinburgh,Scotland,UK,1999.518-529.
[5] Weiss Y,Torralba A,Fergus R.Spectral hashing[A].Proceedings of the Conference on Neural Information Processing Systems[C].Vancouver,Canada,2008.1753-1760.
[6] Gong Y,Lazebnik S,Gordo A,et al.Iterative quantization:a Procrustean approach to learning binary codes for large-scale image retrieval[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2013,35(12):2916-2929.
[7] Kulis B,Darrell T.Learning to Hash with Binary Reconstructive Embeddings[A].Advances in Neural Information Processing Systems[C].Vancouver,Canada,2009.1042-1052.
[8] Liu W,Wang J,Ji R,et al.Supervised hashingwith kernels[A].Proceedings of the Computer Vision and Pattern Recognition[C].Providence,RI,USA:IEEE,2012.2074-2081.
[9] Datar M,Immorlica N,Indyk,et al.Locality-sensitive hashing scheme based on p-stable distributions[A].Proceedings of the Twentieth Annual Symposium on Computational Geometry[C].Brooklyn,NewYork,USA:ACM,2004.253-262.
[10] Kulis B,Grauman K.Kernelized locality sensitive hashing for scalable image search[A].Proceedings of the 12th International Conference on Computer Vision[C].Kyoto,Japan:IEEE,2009.2130-2137.
[11] Kulis B,Jain P,Grauman K.Fastsimilarity search for learned metrics[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2009,31(12):2143-2157.
[12] Mu Y,Yan S.Non-metric locality-sensitive Hashing[A].Proceedings of the 24th AAAI Conference on Artificial Intelligence[C].Atlanta,Georgia,USA,2010.2235-2246.
[13] Oliva A,Torralba A.Building the gist of a scene:the role of global image features in recognition[J].Progress in Brain Research,2006,155(2):23-36.
[14] Csurka G,Dance C R,Fan L,et al.Visual categorization with bags of keypoint[J].Workshop on Statistical Learning in Computer Vision ECCV,2004,44(247):1-22.
[15] Sivic J,Zisserman A.Video google:A text retrieval approach to object matching in videos[A].Proceedings of the 9th IEEE International Conference on Computer Vision[C].Nice,France:IEEE,2003.1470-1477.
[16] Krizhevsky A,Sutskever I,Hinton G E.Image-Net classification with deep convolutional neural networks[A].Proceedings of the International Conference on Neural Information Processing Systems[C].Lake Tahoe,USA,2012.1097-1105.
[17] Babenko A,Slesarev A,Chigorin A,et al.Neu-ral Codes for Image Retrieval[A].Proceedings of the European Conference Computer Vision[C].Zürich,Switzerland:ECCV,2014.584-599.
[18] Taigman Y,Yang M,Ranzato M,et al.DeepFa-ce:Closing the gap to human-level performancein face verification[A].Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition[C].Columbus,OH,USA:IEEE,2014.1701-1708.
[19] Sun Y,Wang X,Tang X.Deep learning face representation from predicting 10,000 classes[A].Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition[C].Columbus,OH,USA:IEEE,2014.1891-1898.
[20] Toshev A,Szegedy C.DeepPose:Human pose estimation via deep neural networks[A].Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition[C].Columbus,OH,USA:IEEE,2014.1653-1660.
[21] Xia R,Pan Y,Lai H,et al.Supervised hashingfor image retrieval via image representation learning[A].Proceedings of AAAI Conference on Artificial Intelligence[C].Quebec,Canada,2014.2156-2162.
[22] Lai H,Pan Y,Liu Y,et al.Simultaneous feature learning and hash coding with deep neural networks[A].Proceedings of the Computer Vision and Pattern Recognition[C].Boston,MA,USA:IEEE,2015.3270-3278.
[23] Lin K,Yang H F,Hsiao J H,et al.Deep learning of binary hash codes for fast image retrieval[A].Proceedings of the Computer Vision and Pattern Recognition Workshops[C].Boston,MA,USA:IEEE,2015.27-35.
[24] Szegedy C,Liu W,Jia Y,et al.Going deeper with convolutions[A].Proceedings of the ComputerVision and Pattern Recognition[C].Boston,MA,USA:IEEE,2015.1-9.
[25] He K,Zhang X,Ren S,et al.Deep residual learning for image recognition[A].Proceedings ofthe Computer Vision and Pattern Recognition[C].Las Vegas,Nevada,USA:IEEE,2016.770-778.
[26] Chua T S,Tang J,Hong R,et al.NUS-WIDE:a real-world web image database from National University of Singapore[A].ACM International Conference on Image and Video Retrieval[C].SantorinIsland,Greece:ACM,2009.48.

一种改进的二进制哈希编码方法

An Improved Binary Hash Coding Method

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	郑兴华, 孙喜庆, 吕嘉欣, 鲜征征, 李磊. 基于深度学习和智能规划的行为识别[J]. 电子学报, 2019, 47(8): 1661-1668.
[2]	查雄, 许漫坤, 彭华, 秦鑫, 李天昀. 基于深度残差网络的特定协议信号识别[J]. 电子学报, 2019, 47(7): 1532-1537.
[3]	刘颖, 胡丹, 范九伦, 王富平, 李大湘. 基于融合特征的现勘图像检索结果优化算法[J]. 电子学报, 2019, 47(2): 296-301.
[4]	张峰, 钟宝江. 基于兴趣目标的图像检索[J]. 电子学报, 2018, 46(8): 1915-1923.
[5]	刘颖, 胡丹, 范九伦. 现勘图像检索综述[J]. 电子学报, 2018, 46(3): 761-768.
[6]	李来, 刘光灿, 孙玉宝, 刘青山. 各向同性的迭代量化哈希算法[J]. 电子学报, 2017, 45(7): 1707-1714.
[7]	毕晓君, 潘铁文. 基于教与学优化算法的相关反馈图像检索[J]. 电子学报, 2017, 45(7): 1668-1676.
[8]	郑云飞, 张雄伟, 曹铁勇, 孙蒙. 基于全卷积网络的语义显著性区域检测方法研究[J]. 电子学报, 2017, 45(11): 2593-2601.
[9]	柯圣财, 赵永威, 李弼程, 彭天强. 基于卷积神经网络和监督核哈希的图像检索方法[J]. 电子学报, 2017, 45(1): 157-163.
[10]	周燕, 曾凡智. 基于二维压缩感知和分层特征的图像检索算法[J]. 电子学报, 2016, 44(2): 453-460.
[11]	赵宏伟, 李清亮, 刘萍萍. 基于分级显著信息的空间编码方法[J]. 电子学报, 2014, 42(9): 1863-1867.
[12]	张瑞杰, 李弼程, 魏福山. 基于多尺度上下文语义信息的图像场景分类算法[J]. 电子学报, 2014, 42(4): 646-652.
[13]	周燕, 曾凡智, 赵慧民, 卢炎生, 周月霞. 一种基于精细化稀疏自适应匹配追踪算法的图像检索方法研究[J]. 电子学报, 2014, 42(12): 2457-2466.
[14]	王斌. 一种基于多尺度拱高形状描述的图像检索方法[J]. 电子学报, 2013, 41(9): 1821-1825.
[15]	陈慧中;陈永光;景宁;陈荦;王钧. 基于显著区域的月球影像内容特征研究[J]. 电子学报, 2012, 40(5): 911-919.