联合加权聚合深度卷积特征的图像检索方法

时璇 1; 许林松 1; 李晨 1; 王佳星 1; 李党超 2

doi:10.7652/xjtuxb201902017

您当前的位置：

首页 >

文章列表页 >

联合加权聚合深度卷积特征的图像检索方法

更新时间：2025-07-09

- 联合加权聚合深度卷积特征的图像检索方法
- Joint Weighting Aggregation of Deep Convolutional Features for Image Retrieval
- 西安交通大学学报 2019年53卷第2期页码：128-135
- 作者机构：
  
  西安交通大学软件学院,西安,710049
- 作者简介：
- 基金信息：
  
  : 2018-08-13。作者简介: 时璇(1993—),女,硕士生;李晨(通信作者),女,讲师,硕士生导师。基金项目: 国家自然科学基金资助项目(61603289)
- DOI：10.7652/xjtuxb201902017
  中图分类号： TP391
- 网络首发：2019-02-10，
  
  纸质出版：2019
- 稿件说明：
移动端阅览
时璇 1, 许林松 1, 李晨 1, 等. 联合加权聚合深度卷积特征的图像检索方法[J]. 西安交通大学学报, 2019,53(2):128-135.

Joint Weighting Aggregation of Deep Convolutional Features for Image Retrieval[J]. 2019, 53(2): 128-135.
时璇 1, 许林松 1, 李晨 1, 等. 联合加权聚合深度卷积特征的图像检索方法[J]. 西安交通大学学报, 2019,53(2):128-135. DOI： 10.7652/xjtuxb201902017.

Joint Weighting Aggregation of Deep Convolutional Features for Image Retrieval[J]. 2019, 53(2): 128-135. DOI： 10.7652/xjtuxb201902017.

摘要

针对图像特征提取不充分影响图像检索平均精确率的问题

提出了一种基于联合加权聚合深度卷积特征的图像检索方法。该方法将图像输入到预先训练好的卷积神经网络中

提取最后一个卷积层输出作为图像的深度卷积特征; 通过计算空间权重矩阵突出图像的显著性区域并抑制背景噪声区域

然后根据通道方差最大原则选取相应的特征图计算出空间权重矩阵

将原始深度卷积特征加权聚合为列向量; 通过区分性地对待不同通道的特征图

计算出通道权重向量与上述列向量点乘得到最终的全局特征向量。公开数据集上的实验结果表明

本文方法能够有效地增强图像特征的表达能力

在图像检索的平均精确率上优于其他同类方法

可以有效地应用到图像检索相关领域。

Abstract

An image retrieval method based on deep convolutional features of joint weighting aggregation is proposed to solve the problem that most existing image retrieval approaches can't fully extract image features and their performance requires to be improved. Firstly

the method extracts the outputs of the last convolutional layer as deep convolutional features of an image by passing the image through a pre-trained deep convolutional neural network. Then

the spatial weight matrix is calculated to highlight significance regions of the image and to suppress the background noise of the image. The maximum-principle of channel variance is then used to select the corresponding feature map and to calculate the spatial weight matrix. The original deep convolutional features are weighted and aggregated into a feature vector. Moreover

the channel weight vector is calculated by distinguishing feature maps of different channels

and then the global feature representation of this image is obtained by multiplying the aggregated feature vector and the channel weight. Experimental results on different public available datasets for image retrieval show that the proposed approach effectively enhances the discriminative ability of image features

outperforms the state-of-the-art approaches based on pre-trained networks and can be effectively applied to related fields of image retrieval.

关键词

Keywords

references

徐思雨, 蔡佳妮, 祝继华, 等. 自适应多位编码量化的哈希图像检索方法 [J]. 西安交通大学学报, 2017, 51(8): 19-25.

XU Siyu, CAI Jiani, ZHU Jihua, et al. An adaptive hashing retrieval method of images based on multi-bit quantization [J]. Journal of Xi'an Jiaotong University, 2017, 51(8): 19-25.

LOWE D G. Distinctive image features from scale-invariant keypoints [J]. International Journal of Computer Vision, 2004, 60(2): 91-110.

JEGOU H, PERRONNIN F, DOUZE M, et al. Aggregating local image descriptors into compact codes [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012, 34(9): 1704-1716.

JEGOU H, ZISSERMAN A. Triangulation embedding and democratic aggregation for image search [C]∥Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ, USA: IEEE, 2014: 3310-3317.

JEGOU H, DOUZE M, SCHMID C, et al. Aggregating local descriptors into a compact image representation [C]∥Proceedings of the 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Piscataway, NJ, USA: IEEE, 2010: 3304-3311.

PERRONNIN F, LIU Y, SANCHEZ J, et al. Large-scale image retrieval with compressed Fisher vectors [J]. Computer Vision and Pattern Recognition, 2010, 26(2): 3384-3391.

KRIZHEVSKY A, SUTSKEVER I, HINTON G E. ImageNet classification with deep convolutional neural networks [C]∥Proceedings of the International Conference on Neural Information Processing Systems. Cambridge, MA, USA: MIT Press, 2012: 1097-1105.

WEI Xiushen, LUO Jianhao, WU Jianxin, et al. Selective convolutional descriptor aggregation for fine-grained image retrieval [J]. IEEE Transactions on Image Processing, 2016, 26(6): 2868-2881.

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.

ZHENG Liang, YANG Yi, TIAN Qi. SIFT meets CNN: a decade survey of instance retrieval [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018, 40(5): 1224-1244.

BABENKO A, SLESAREV A, CHIGORIN A, et al. Neural codes for image retrieval [C]∥Proceedings of the 13th European Conference on Computer Vision. Berlin, Germany: Springer, 2014: 584-599.

YANDEX A B, LEMPITSKY V. Aggregating local deep features for image retrieval [C]∥Proceedings of the 2016 IEEE International Conference on Computer Vision. Piscataway, NJ, USA: IEEE, 2016: 1269-1277.

KALANTIDIS Y, MELLINA C, OSINDERO S. Cross-dimensional weighting for aggregated deep convolutional features [C]∥Proceedings of the 14th European Conference on Computer Vision. Berlin, Germany: Springer-Verlag, 2016: 685-701.

TOLIAS G, SICRE R, JÉGOU H. Particular object retrieval with integral max-pooling of CNN activations[EB/OL].(2016-02-24)[2017-08-20]. https: ∥arxiv. org/abs/1511.05879v2.

WANG Jiaxing, ZHU Jihua, PANG Shanming, et al. Adaptive co-weighting deep convolutional features for object retrieval[EB/OL].(2018-03-20)[2018-04-20]. https: ∥arxiv. org/abs/1803.07360.

JEGOU H, DOUZE M, SCHMID C. On the burstiness of visual elements [C]∥Proceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ, USA: IEEE, 2009: 1169-1176.

XU Jian, WANG Chunheng, QI Chengzuo, et al. Unsupervised semantic-based aggregation of deep convolutional features [J]. IEEE Transactions on Image Processing, 2018, 28(2): 601-611.

SIMONYAN K, ZISSERMAN A. Very deep convolutional networks for large-scale image recognition[EB/OL].(2015-04-10)[2017-07-20]. https: ∥arxiv. org/abs/1409.1556.

PHILBIN J, CHUM O, ISARD M, et al. Object retrieval with large vocabularies and fast spatial matching [C]∥Proceedings of the 2007 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ, USA: IEEE, 2007: 1-8.

JAMES P, ONDREJ C, MICHAEL I, et al. Lost in quantization: improving particular object retrieval in large scale image databases [C]∥Proceedings of the 26th IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ, USA: IEEE, 2008: 1-8.

JEGOU H, DOUZE M, SCHMID C. Improving bag-of-features for large scale image search [J]. International Journal of Computer Vision, 2010, 87(3): 316-336.

ARANDJELOVIC R, GRONAT P, TORII A, et al. NetVLAD: CNN architecture for weakly supervised place recognition [C]∥Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ, USA: IEEE, 2016: 5297-5307.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

采用类别相似度聚合的关联文本分类方法

一种利用局部结构信息的加权哈希图像检索算法

一种判别极端学习的相关反馈图像检索方法

一种三结构描述子的图像检索方法

采用试验设计方法的第三代CO2气体箔轴承箔片几何参数敏感性分析