一种新的基于SVM和主动学习的图像检索方法

J4 ›› 2014, Vol. 36 ›› Issue (07): 1371-1376.

一种新的基于SVM和主动学习的图像检索方法

彭晏飞1，尚永刚1,王德建2

(1.辽宁工程技术大学电子与信息工程学院，辽宁葫芦岛 125105;2.渤海装备辽河重工有限公司，辽宁盘锦 124010)

收稿日期:2012-12-18 修回日期:2013-05-10 出版日期:2014-07-25 发布日期:2014-07-25
基金资助:
国家自然科学基金资助项目（61172144）

A novel image retrieval method
based on SVM and active learning

PENG Yanfei1,SHANG Yonggang1,WANG Dejian2

(1.School of Electronics and Information Engineering,Liaoning Technical University,Huludao 125105;
2.China Detroleum Liaohe Equipment Company,Panjin 124010,China)

Received:2012-12-18 Revised:2013-05-10 Online:2014-07-25 Published:2014-07-25

摘要/Abstract

摘要：

在基于内容的图像检索中，支持向量机（SVM）能够很好地解决小样本问题，而主动学习算法则可以根据学习进程主动选择最佳的样本进行学习，大幅度缩短训练时间，提高分类算法效率。为使图像检索更加快速、高效，提出一种新的基于SVM和主动学习的图像检索方法。该方法根据SVM构造分类器，通过“V”型删除法快速缩减样本集，同时通过最优选择法从缩减样本集中选取最优的样本作为训练样本，最终构造出不仅信息度大而且冗余度低的最优训练样本集，从而训练出更好的SVM分类器，得到更高的检索效率。实验结果表明，与传统的SVM主动学习的图像检索方法相比，该方法能够较大幅度提高检索性能。

关键词: 图像检索, 支持向量机, 主动学习, &ldquo, V&rdquo, 型删除法, 最优选择法

Abstract:

In the content-based image retrieval, Support Vector Machine (SVM) can resolve the problem of small sample size, and the active learning algorithm can select the most optimal samples to learn actively according to the learning process, thus reducing the training time greatly and improving the efficiency of the classification algorithm. In order to obtain the more rapid and efficient image retrieval, a novel image retrieval method based on SVM and active learning is proposed. Firstly, the method constructs the classifier on the basis of SVM. Secondly, the “V” elimination method is used to reduce the sample sets quickly, and the optimal selection method is applied to select the optimal samples from the reduced sample sets as the training ones. Finally, the optimal training sample set with abundant information and lower redundancy is obtained, so that the better SVM based classifier is constructed and the higher retrieval efficiency is achieved. Experimental results show that, compared with the traditional image retrieval method based on SVM and active learning, the proposed method has better performance and can improve the retrieval performance greatly.

Key words: image retrieval;SVM;active learning;“V”elimination method;optimal selection method

彭晏飞1，尚永刚1,王德建2. 一种新的基于SVM和主动学习的图像检索方法[J]. J4, 2014, 36(07): 1371-1376.

PENG Yanfei1,SHANG Yonggang1,WANG Dejian2. A novel image retrieval method
based on SVM and active learning [J]. J4, 2014, 36(07): 1371-1376.

参考文献

［1］Wu Hong, Lu Hanqing, Ma Songde. A survey of relevance feedback techniques in contentbased image retrieval ［J］. Chinese Journal of Computers, 2005, 28(12):19691979.(in Chinese)
［2］Li Xiangyang,Zhuang Yueting,Pan Yunhe.The technique and systems of contentbased image retrieval ［J］. Journal of Computer Research and Development, 2001, 38(3):344352.(in Chinese)
［3］Haykin S. Neural networks and learning machines［M］. 3rd ed. MA:PrenticeHall, 2008.
［4］Zhang Lei, Lin Fuzong, Zhang Bo. Support vector machine learning for image retrieval ［C］∥Proc of IEEE International Conference on Image Processing, 2001:721724.
［5］Long Jun, Yin Jianping, Zhu En, et al. A survey of active learning ［J］. Journal of Computer Research and Development, 2008,45（Suppl）:300304. (in Chinese)
［6］Wu Weining, Guo Maozu, Liu Yang. A method of active learning with optimal sampling strategy ［C］∥Proc of CSAE’12, 2012:725729.
［7］Xie HongSheng,Zhang Hong.Active learning method based on support vector machine in contentbased image retrieval ［J］. Journal of Shandong Normal University(Natural Science), 2007,22(4)：4648.(in Chinese)
［8］Tong S, Chang E. Support vector machine active learning for image retrieval ［C］∥Proc of the 9th ACM International Conference on Multimedia, 2001:107119.
［9］Zhang Yufang, Chen Zhuo, Xiong Zhongyang, et al. Image retrieval method based on SVMs and active learning ［J］. Computer Engineering and Applications, 2010, 46(24):193196.(in Chinese)
［10］Cohn D A, Atlas L, Ladner R E. Improving generalization with active learning［J］. Machine Learning, 1994,15(2):201221.
［11］Seung H S, Opper M, Sompolinsky H. Query by committee［C］∥Proc of the 5th Annual ACM Conference on Computational Learning Theory, 1992:287294.
［12］Zhu Fang, Gu Junhua, Yang Xinwei, et al. New reduction strategy of largescale training sample set for SVM ［J］. Journal of Computer Applications, 2009,29(10):27362740.(in Chinese)
［13］Feng Guohe.Research on large scale SVM classification based on boundary Knearest ［J］. Computer Engineering and Applications, 2009,45（23）:1517.(in Chinese)
附中文参考文献：
［1］吴洪, 卢汉清, 马颂德. 基于内容图像检索中相关训练技术的回顾［J］. 计算机学报, 2005, 28(12):19691979.
［2］李向阳, 庄越挺, 潘云鹤. 基于内容的图像检索技术与系统［J］. 计算机研究与发展, 2001, 38(3):344352.
［5］龙军，殷建平，祝恩，等. 主动学习研究综述［J］. 计算机研究与发展，2008,45（Suppl）:300304.
［7］解洪胜，张虹. 基于支持向量机的图像检索主动学习方法［J］. 山东师范大学学报（自然科学版），2007,22(4)：4648.
［9］张玉芳, 陈卓, 熊忠阳，等. 一种基于SVM和主动学习的图像检索方法［J］. 计算机工程与应用,2010, 46(24):193196.
［12］朱方，顾军华，杨欣伟，等. 一种新的支持向量机大规模训练样本集缩减策略［J］.计算机应用，2009,29(10):27362740.
［13］奉国和. 边界K邻近大样本支持向量机分类［J］.计算机工程与应用，2009,45（23）：1517.

[1]	刘屹成, 刘晓燕, 严馨. 并行平衡级联支持向量机[J]. 计算机工程与科学, 2023, 45(07): 1170-1177.
[2]	唐宇, 代琪, 杨梦园, 陈丽芳, . 改进麻雀搜索算法优化SVM的异常点检测[J]. 计算机工程与科学, 2023, 45(02): 346-354.
[3]	仲臣, 余学祥, 邰晓曼, 韩雨辰, 肖星星, 刘清华, . 萤火虫算法优化支持向量机室内定位研究[J]. 计算机工程与科学, 2022, 44(11): 1968-1975.
[4]	强梓林, 刘建国, 刘云峰, 卫栋, 强彦. 基于时域-频域哈希编码的电网图像检索方法[J]. 计算机工程与科学, 2022, 44(10): 1877-1884.
[5]	马汉达, 朱敏. 改进SVM不平衡数据分类的IGWOSMOTE方法[J]. 计算机工程与科学, 2022, 44(06): 1133-1140.
[6]	姬子恒, 王斌. 基于深度学习的草图检索方法研究进展[J]. 计算机工程与科学, 2021, 43(12): 2190-2205.
[7]	张晗,宋志华,王彤,朱良谊. 基于军事背景的大学计算机课程案例教学探索与实践[J]. 计算机工程与科学, 2019, 41(增刊S1): 210-212.
[8]	贺周雨1,冯旭鹏2,刘利军1,黄青松1,3. 基于SHN模型的商品图像检索方法[J]. 计算机工程与科学, 2019, 41(11): 1991-1999.
[9]	翟俊海1,2，齐家兴1,2,沈矗1,2,宋丹丹1,2,王谟瀚1,2,田石1,2. 基于MapReduce和Spark的大数据主动学习比较研究[J]. 计算机工程与科学, 2019, 41(10): 1715-1722.
[10]	韦修喜1，黄华娟1，周永权1,2. 基于AP聚类的约简孪生支持向量机快速分类算法[J]. 计算机工程与科学, 2019, 41(10): 1899-1904.
[11]	张诗情，杨耀华，沈立，王志英. 通过部分页迁移实现CPU-GPU高效透明的数据通信[J]. 计算机工程与科学, 2019, 41(07): 1168-1175.
[12]	王云艳1,2，罗冷坤1，王重阳1. 基于流形学习的光学遥感图像分类[J]. 计算机工程与科学, 2019, 41(07): 1212-1219.
[13]	冯璋，裴东,王维. 基于改进灰狼算法优化支持向量机的人脸识别[J]. 计算机工程与科学, 2019, 41(06): 1057-1063.
[14]	袁晖1，廖开阳1,3，郑元林1,2，曹从军1,3，汤梓伟1，邓轩1. 基于CNN特征加权和区域整合的图像检索[J]. 计算机工程与科学, 2019, 41(01): 113-121.
[15]	崔建双，车梦然. 基于多分类支持向量机的优化算法智能推荐系统与实证分析[J]. 计算机工程与科学, 2019, 41(01): 153-160.