相似性保持实例检索方法

doi:10.13229/j.cnki.jdxbgxb20181057

吉林大学学报(工学版) ›› 2019, Vol. 49 ›› Issue (6): 2045-2050.doi: 10.13229/j.cnki.jdxbgxb20181057

相似性保持实例检索方法

赵宏伟^1,²(),王鹏¹,范丽丽¹,胡黄水³,刘萍萍¹()

1. 吉林大学计算机科学与技术学院，长春 130012
2. 中国科学院应用光学国家重点实验室，长春 130033
3. 长春工业大学计算机科学与工程学院，长春 130012

收稿日期:2018-10-22 出版日期:2019-11-01 发布日期:2019-11-08
通讯作者: 刘萍萍 E-mail:zhaohw@jlu.edu.cn;liupp@jlu.edu.cn
作者简介:赵宏伟(1962-),男,教授，博士生导师.研究方向:嵌入式人工智能.E-mail:zhaohw@jlu.edu.cn
基金资助:
吉林省省级科技创新专项资金项目(20190302026GX);吉林省发改委产业技术研究与开发项目(2019C054-4);应用光学国家重点实验室开放基金项目(20173660)

Similarity retention instance retrieval method

Hong-wei ZHAO^1,²(),Peng WANG¹,Li-li FAN¹,Huang-shui HU³,Ping-ping LIU¹()

1. College of Computer Science and Technology，Jilin University，Changchun 130012，China
2. State Key Laboratory of Applied Optics，Chinese Academy of Science，Changchun 130033，China
3. School of Computer Science and Engineering，Changchun University of Technology，Changchun 130012，China

Received:2018-10-22 Online:2019-11-01 Published:2019-11-08
Contact: Ping-ping LIU E-mail:zhaohw@jlu.edu.cn;liupp@jlu.edu.cn

摘要/Abstract

摘要：

针对多输入样本的网络训练，提出了相似性保持实例检索方法。首先，经深度网络除卷积结构提取输入图像的特征并加以池化操作。然后，依照基准排序，修正低相关性图像与查询图像的相似度关系，获得低相关性图像的对比损失系数，保留损失基准值内的损失值，进行基于相似性保持的对比损失训练。最后，使用训练后网络提取图像特征，用于进行实例级图像检索。实验结果表明：基于相似性保持的损失对比函数具有可行性，该方法显著提高了实例级图像的检索精确度。

关键词: 计算机应用, 图像检索, 卷积神经网络, 损失函数, 排序学习, 相似性保持

Abstract:

Aiming at the network training of multi-input samples, a similarity retention instance retrieval method is proposed. Firstly, the input image features are extracted by the convolution structure in the depth network and pooled. Then, according to the benchmark order, the similarity relationship between the low correlation image and the query image is corrected, and the low correlation image contrast loss coefficient is obtained, and the loss value within the loss reference value is retained. The loss value is performed to maintain contrast loss training based on similarity. Finally, the post-training network is used to extract image features for instance-level image retrieval. The experimental results show that the loss comparison function based on similarity is feasible, and the method significantly improves the accuracy of instance-level image retrieval.

Key words: computer application, image retrieval, convolutional neural network(CNN), loss function, sorting learning, similarity retention

中图分类号:

TP391

赵宏伟,王鹏,范丽丽,胡黄水,刘萍萍. 相似性保持实例检索方法[J]. 吉林大学学报(工学版), 2019, 49(6): 2045-2050.

Hong-wei ZHAO,Peng WANG,Li-li FAN,Huang-shui HU,Ping-ping LIU. Similarity retention instance retrieval method[J]. Journal of Jilin University(Engineering and Technology Edition), 2019, 49(6): 2045-2050.

图/表 8

图1

图2

图3

图4

表1

图5

表2

表3

参考文献

1	王方石, 王坚, 李兵, 等 . 基于深度属性学习的交通标志检测[J]. 吉林大学学报：工学版, 2018, 48(1): 319-329.
1	Wang Fang-shi , Wang Jian , Li Bing , et al . Deep attribute learning based traffic sign detection[J]. Journal of Jilin University (Engineering and Technology Edition), 2018, 48(1): 319-329.
2	李琳辉，伦智梅，连静，等 . 基于卷积神经网络的道路车辆检测方法[J]. 吉林大学学报：工学版，2017，47(2):384-391.
2	Li Lin-hui , Zhi-mei Lun , Lian Jing , et al . Convolution neural network-based vehicle detection method[J]. Journal of Jilin University (Engineering and Technology Edition), 2017, 47(2): 384-391.
3	Wang Zhi-peng , Xiang Xuan-lu , Zhao Zhi-cheng , et al . Deep image retrieval: indicator and gram matrix weighting for aggregated convolutional features[C]∥2018 IEEE International Conference on Multimedia and Expo (ICME), San Diego, CA, USA， 2018: 1-6.
4	Xia Zhi-hua , Wang Xin-hui , Zhang Lian-gao , et al . A privacy-preserving and copy-deterrence content-based image retrieval scheme in cloud computing[J]. IEEE Transactions on Information Forensics & Security, 2016, 11(11): 2594-2608.
5	刘富, 宗宇轩, 康冰, 等 . 基于优化纹理特征的手背静脉识别系统[J]. 吉林大学学报：工学版, 2018, 48(6): 1844-1850.
5	Liu Fu , Zong Yu-xuan , Kang Bing , et al . Dorsal hand vein recognition system based on optimized texture features[J]. Journal of Jilin University (Engineering and Technology Edition), 2018, 48(6): 1844-1850.
6	Wang J , Song Y , Leung T , et al . Learning fine-grained image similarity with deep ranking[C]∥Computer Vision and Pattern Recognition, Columbus, OH, 2014: 1386-1393.
7	Hadsell R , Chopra S , LeCun Y . Dimensionality reduction by learning an invariant mapping[C]∥IEEE Computer Society Conference on Computer Vision and Pattern Recognition, New York, NY, USA, 2006: 1735-1742.
8	Radenovi? F , Tolias G , Chum O . Fine-tuning CNN image retrieval with no human annotation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2019, 41(7): 1655-1668.
9	Krizhevsky A , Sutskever I , Hinton G E . Imagenet classification with deep convolutional neural networks[J/OL]. [2018-10-10].https:∥iphysresearch.github.io/paper_summary/ImageNet%20Classification%20with%20Deep%20Convolutional%20Neural%20Networks.pdf.
10	Simonyan K , Zisserman A . Very deep convolutional networks for large-scale image recognition[J/OL].[2018-10-10]. https:∥.
11	He K , Zhang X , Ren S , et al . Deep residual learning for image recognition[C]∥IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA, 2016: 770-778.
12	Razavian A S , Sullivan J , Carlsson S , et al . Visual instance retrieval with deep convolutional networks[J]. ITE Transactions on Media Technology and Applications, 2016, 4(3): 251-258.
13	Babenko A , Lempitsky V . Aggregating deep convolutional features for image retrieval[J/OL].[2018-10-11].https:∥arxiv.org/pdf/ 1510.07493.pdf.
14	Luukka P , Lepp?lampi T . Similarity classifier with generalized mean applied to medical data[J]. Computers in Biology and Medicine, 2006, 36(9): 1026-1040.
15	Philbin J , Chum O , Isard M , et al . Lost in quantization: Improving particular object retrieval in large scale image databases[J/OL]. [2018-10-15].http:∥.
16	Kingma D P , Ba J l . Adam: a method for stochastic optimization[J/OL].[2018-10-11]. https:∥arxiv.org/pdf/ 1412.6980.pdf.
17	Donahue J , Jia Y , Vinyals O , et al . DeCAF: a deep convolutional activation feature for generic visual recognition[J]. International Conference on Machine Learning, 2013, 32: 647-655.
18	Philbin J , Chum O , Isard M , et al . Object retrieval with large vocabularies and fast spatial matching[C]∥2007 IEEE Conference on Computer Vision and Pattern Recognition, Minneapolis, MN, USA, 2007: 9738061.
19	Jégou H , Zisserman A . Triangulation embedding and democratic aggregation for image search[J/OL].[2018-10-09]. https:∥.
20	Babenko A , Slesarev A , Chigorin A , et al . Neural codes for image retrieval[J/OL].[2018-10-09]. https:∥arxiv.org/pdf/ 1404.1777.pdf.
21	Sharif R A , Sullivan J , Maki A , et al . A baseline for visual instance retrieval with deep convolutional networks[C]∥International Conference on Learning Representations, San Diego, CA, 2015:165765.
22	Ng J Y H , Yang F , Davis L S . Exploiting local features from deep networks for image retrieval[C]∥Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, Boston, MA, USA, 2015: 53-61.
23	Paulin M , Douze M , Harchaoui Z , et al . Local convolutional features with unsupervised training for image retrieval[C]∥Proceedings of the IEEE International Conference on Computer vision, Santiago, Chile, 2015: 91-99.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

池化方法	初始k	数据集
池化方法	初始k	Oxford5k	Paris6k
MAC	正无穷	54.93	66.06
SPoC	1	45.63	66.46
GeM	3	57.95	69.29

方法	特征维度	数据集
方法	特征维度	Oxford5k	Paris6k
文献[19]	1024	56.0	-
文献[19]	128	43.3	-
文献[20]	128	55.7	-
文献[21]	256	53.3	67.0
文献[13]	256	53.1	-
文献[22]	128	59.3	59.0
文献[23]	256	56.5	-
本文	256	66.15	71.78

[1]	沈军,周晓,吉祖勤. 服务动态扩展网络及其结点系统模型的实现[J]. 吉林大学学报(工学版), 2019, 49(6): 2058-2068.
[2]	车翔玖,刘华罗,邵庆彬. 基于Fast RCNN改进的布匹瑕疵识别算法[J]. 吉林大学学报(工学版), 2019, 49(6): 2038-2044.
[3]	周柚,杨森,李大琳,吴春国,王岩,王康平. 基于现场可编程门电路的人脸检测识别加速平台[J]. 吉林大学学报(工学版), 2019, 49(6): 2051-2057.
[4]	周炳海,吴琼. 考虑工具和空间约束的机器人装配线平衡优化[J]. 吉林大学学报(工学版), 2019, 49(6): 2069-2075.
[5]	马子骥,卢浩,董艳茹. 双通道单图像超分辨率卷积神经网络[J]. 吉林大学学报(工学版), 2019, 49(6): 2089-2097.
[6]	李宾,周旭,梅芳,潘帅宁. 基于K-means和矩阵分解的位置推荐算法[J]. 吉林大学学报(工学版), 2019, 49(5): 1653-1660.
[7]	郭继昌,吴洁,郭春乐,朱明辉. 基于残差连接卷积神经网络的图像超分辨率重构[J]. 吉林大学学报(工学版), 2019, 49(5): 1726-1734.
[8]	孙延君,申铉京,陈海鹏,赵永哲. 基于局部平面线性点的翻拍图像鉴别算法[J]. 吉林大学学报(工学版), 2019, 49(4): 1320-1328.
[9]	李雄飞,宋璐,张小利. 基于协同经验小波变换的遥感图像融合[J]. 吉林大学学报(工学版), 2019, 49(4): 1307-1319.
[10]	翟凤文,党建武,王阳萍,金静,罗维薇. 基于扩展轮廓的快速仿射不变特征提取[J]. 吉林大学学报(工学版), 2019, 49(4): 1345-1356.
[11]	刘元宁,刘帅,朱晓冬,霍光,丁通,张阔,姜雪,郭书君,张齐贤. 基于决策粒子群优化与稳定纹理的虹膜二次识别[J]. 吉林大学学报(工学版), 2019, 49(4): 1329-1338.
[12]	李宾,申国君,孙庚,郑婷婷. 改进的鸡群优化算法[J]. 吉林大学学报(工学版), 2019, 49(4): 1339-1344.
[13]	王楠,李金宝,刘勇,张玉杰,钟颖莉. TPR⁃TF:基于张量分解的时间敏感兴趣点推荐模型[J]. 吉林大学学报(工学版), 2019, 49(3): 920-933.
[14]	刘富,宗宇轩,康冰,张益萌,林彩霞,赵宏伟. 基于优化纹理特征的手背静脉识别系统[J]. 吉林大学学报(工学版), 2018, 48(6): 1844-1850.
[15]	王利民,刘洋,孙铭会,李美慧. 基于Markov blanket的无约束型K阶贝叶斯集成分类模型[J]. 吉林大学学报(工学版), 2018, 48(6): 1851-1858.

相似性保持实例检索方法

Similarity retention instance retrieval method

RICH HTML

PDF (PC)

摘要/Abstract

引用本文

使用本文

图/表 8

参考文献

相关文章 15

Metrics

本文评价

推荐阅读 10