基于Bag-of-words和Hash编码的近似重复图像检测算法

doi:10.3724/SP.J.1087.2013.00667

计算机应用 ›› 2013, Vol. 33 ›› Issue (03): 667-669.DOI: 10.3724/SP.J.1087.2013.00667

基于Bag-of-words和Hash编码的近似重复图像检测算法

王誉天^1*,袁江涛²,秦海权¹,刘鑫¹

1.公安部第一研究所, 北京 100048;
2.天津市公安局北辰分局,天津 300400

收稿日期:2012-09-10 修回日期:2012-11-02 出版日期:2013-03-01 发布日期:2013-03-01
通讯作者: 王誉天
作者简介:王誉天(1975-),男,陕西咸阳人,工程师,硕士,主要研究方向:电子与计算机测试; 袁江涛(1976-),男,甘肃天水人,工程师,硕士,主要研究方向:计算机网络; 秦海权(1981-),男,湖南永州人,工程师,硕士,主要研究方向:计算机信息安全、数据鉴定; 刘鑫(1980-),女,山东济宁人,工程师,硕士,主要研究方向:计算机信息安全。

Algorithm of near-duplicate image detection based on Bag-of-words and Hash coding

WANG Yutian^1*, YUAN Jiangtao², QIN Haiquan¹, LIU Xin¹

1.The First Research Institute, Ministry of Public Security, Beijing 100048, China;
2.Beichen Branch, Tianjin Municipal Public Security Bureau, Tianjin 300400, China

Received:2012-09-10 Revised:2012-11-02 Online:2013-03-01 Published:2013-03-01

摘要/Abstract

摘要： 针对近似重复图像检测的传统算法存在检测效率和准确率不够高的缺点,提出了基于Bag-of-words和哈希编码的近似重复图像检测算法。该算法首先利用Bag-of-words把一幅图像表示成一个500维的特征向量; 然后,利用主成分分析(PCA)和尺度不变特征转换(SIFT)进行特征降维,并利用Hash编码技术对特征进行编码; 最后,利用动态距离度量技术实现近似重复图像的检测。实验结果表明,利用该算法进行近似重复图像检测是完全可行的,在准确度和查全率之间做到了较好的平衡,查准率可达90%~95%,查全率可达70%~80%。

关键词: 近似重复图像, Bag-of-words, 主成分分析, 哈希编码, 动态距离度量

Abstract: To solve the low efficiency and precision of the traditional methods, a near-duplicate image detection algorithm based on Bag-of-words and Hash coding was proposed in this paper. Firstly, a 500-dimensional feature vector was used to represent an image by Bag-of-words; secondly, feature dimension was reduced by Principal Component Analysis (PCA) and Scale-Invariant Feature Transform (SIFT) and features were encoded by Hash coding; finally, dynamic distance metric was used to detect near-duplicate images. The experimental results show that the algorithm based on Bag-of-words and Hash coding is feasible in detecting near-duplicate images. This algorithm can achieve a good balance between precision and recall rate: the precision rate can reach 90%-95%, and entire recall rate can reach 70%-80%.

Key words: near-duplicate image, Bag-of-words, Principal Component Analysis (PCA), Hash coding, dynamic distance metric

中图分类号:

TP301.4

王誉天袁江涛秦海权刘鑫. 基于Bag-of-words和Hash编码的近似重复图像检测算法[J]. 计算机应用, 2013, 33(03): 667-669.

WANG Yutian YUAN Jiangtao QIN Haiquan LIU Xin. Algorithm of near-duplicate image detection based on Bag-of-words and Hash coding[J]. Journal of Computer Applications, 2013, 33(03): 667-669.

参考文献

[1]SIVIC J,RUSSELL B,EFROS A A. Discovering objects and their location in images[C]// Tenth IEEE International Conference on Computer Vision. Piscataway, NJ: IEEE Press, 2005:370-377.
[2]HAMPAPUR A, BOLLE R M. Comparison of distance measures for video copy detection [C]// IEEE International Conference on Multimedia and Expo. Piscataway, NJ: IEEE Press, 2001: 737-740.
[3]ZHANG D Q, CHANG S F. Detecting image near-duplicate by stochastic attributed relational graph matching with learning[C]// Proceedings of the 12th Annual ACM MULTIMEDIA ′04. New York: ACM Press, 2004:877-884.
[4]QAMRA A, MENG Y, CHANG E Y. Enhanced perceptual distance functions and indexing for image replicate recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005,27(3):379-391.
[5]PHILBIN C J, ISARD M, ZISSERMAN A. Scalable near identical image and shot detection[C]// Proceedings of the 6th ACM International Conference on Image and Video Retrieval. New York: ACM Press, 2007.
[6]MARET Y, NIKOLOPOULOS S, DUFAUX F, et al. A novel replica detection system using binary classifiers, R-trees, and PCA[C]// International Conference on Image Processing. Piscataway, NJ: IEEE Press, 2006.
[7]QAMRA L, MENG Y, CHANG E. Enhanced perceptual distance functions and indexing for image replica recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005,27(3):379-391.
[8]PHILBIN C J, ZISSERMAN A. Near duplicate image detection: min-Hash and TF-IDF weighting [EB/OL]. [2010-10-10]. http://cmp.felk.cvut.cz/~chum/papers/chum_bmvc08.pdf.
[9]蔺博宇,李弼成,高毫林. 基于贪婪树的外部支持向量机近似重复图像聚类算法[J]. 信号处理, 2012,28(4):601-606.
[10]TANG X. Book retrieval based on near-duplicate image matching[C]// Proceedings of the 9th International Conference on Fuzzy Systems and Knowledge Discovery. Piscataway, NJ: IEEE Press,2012:2616-2619.
[11]WANG G, ZHANG Y, LI F F. Using dependent regions for object categorization in a generative framework[C]// IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Washington, DC: IEEE Computer Society,2006:1597-1604.
[12]SEMPERE V, ALBERO T, SILVESTRE J. Analysis of communication alternatives in a heterogeneous network for a supervision and control system[J].Computer Communications,2006,29(8):1133-l145.
[13]唐坚刚,王泽兴.基于Hash 值的重复图像检测算法[J]. 计算机工程,2009,35(1):183-185.

[1]	王心, 朱浩华, 刘光灿. 卷积鲁棒主成分分析[J]. 计算机应用, 2021, 41(5): 1314-1318.
[2]	陆荣秀, 陈明明, 杨辉, 朱建勇. 基于溶液图像时序特征的元素组分含量动态监测系统[J]. 计算机应用, 2021, 41(10): 3075-3081.
[3]	陈利霞, 班颖, 王学文. 基于张量核范数与3D全变分的背景减除[J]. 计算机应用, 2020, 40(9): 2737-2742.
[4]	郑延斌, 韩梦云, 樊文鑫. 基于二维主成分分析与卷积神经网络的手写体汉字识别[J]. 计算机应用, 2020, 40(8): 2465-2471.
[5]	李东博, 黄铝文. 重加权稀疏主成分分析算法及其在人脸识别中的应用[J]. 计算机应用, 2020, 40(3): 717-722.
[6]	王海鹏, 降爱莲, 李鹏翔. 牛顿-软阈值迭代鲁棒主成分分析算法[J]. 计算机应用, 2020, 40(11): 3133-3138.
[7]	张晓博, 杨燕, 李天瑞, 陆凡, 彭莉兰. 基于医疗文本数据聚类的帕金森病早期诊断预测[J]. 计算机应用, 2020, 40(10): 3088-3094.
[8]	陈莉, 陈晓云. 结合局部熵和鲁棒主成分分析的眼底图像硬性渗出物检测方法[J]. 计算机应用, 2019, 39(7): 2134-2140.
[9]	周非, 夏鹏程. 基于主成分分析和卡方距离的信号强度差指纹定位算法[J]. 计算机应用, 2019, 39(5): 1405-1410.
[10]	陈万志, 徐东升, 张静, 唐雨. 结合优化支持向量机与K-means++的工控系统入侵检测方法[J]. 计算机应用, 2019, 39(4): 1089-1094.
[11]	王鑫, 李可, 徐明君, 宁晨. 改进的基于深度学习的遥感图像分类算法[J]. 计算机应用, 2019, 39(2): 382-387.
[12]	万源, 张景会, 吴克风, 孟晓静. 基于多层非负局部Laplacian稀疏编码的图像分类[J]. 计算机应用, 2018, 38(9): 2489-2494.
[13]	孙莞格, 夏克文, 兰璞. 正则化的加权不完全鲁棒主成分分析方法及其在无线传感器网络节点轨迹拟合中的应用[J]. 计算机应用, 2018, 38(6): 1709-1714.
[14]	范君, 王新, 徐慧. 粒子群优化混合核极限学习机的构造煤厚度预测方法[J]. 计算机应用, 2018, 38(6): 1820-1825.
[15]	唐家琪, 吴璟莉. 基于PPI网络与机器学习的蛋白质功能预测方法[J]. 计算机应用, 2018, 38(3): 722-727.

基于Bag-of-words和Hash编码的近似重复图像检测算法

Algorithm of near-duplicate image detection based on Bag-of-words and Hash coding

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