计算机工程 ›› 2019, Vol. 45 ›› Issue (6): 230-236.doi: 10.19678/j.issn.1000-3428.0052565

• 人工智能及识别技术 • 上一篇    下一篇

联合耦合字典学习与图像正则化的跨媒体检索方法

刘芸1,于治楼2,付强1   

  1. 1.山东师范大学 信息科学与工程学院,济南 250358; 2.浪潮集团有限公司,济南 250101
  • 收稿日期:2018-09-04 出版日期:2019-06-15 发布日期:2019-06-15
  • 作者简介:刘芸(1992—),女,硕士研究生,主研方向为跨媒体检索、机器学习;于治楼(通信作者),研究员;付强,硕士研究生。
  • 基金项目:

    国家自然科学基金(61373081)。

Cross-media retrieval method fusing with coupled dictionary learning and image regularization

LIU Yun1,YU Zhilou2,FU Qiang1   

  1. 1.School of Information Science and Engineering,Shandong Normal University,Jinan 250358,China;2.Inspur Group Co.,Ltd.,Jinan 250101,China
  • Received:2018-09-04 Online:2019-06-15 Published:2019-06-15

摘要:

跨媒体检索方法多数将2个模态的原始特征映射到公共子空间,在子空间中执行跨媒体检索,忽略了判别特征的选择以及模态间的关系。为此,提出一种基于耦合字典学习和图形正则化的新型跨模态检索方法。通过关联和联合更新不同模态的字典,为不同的模态生成均匀的稀疏表示。将不同模态的稀疏表示投影到由类标签信息定义的公共子空间中,以执行跨模态匹配,同时对投影矩阵施加21范数项,选择特征空间的相关和辨别性特征。在此基础上,利用图正则化项保留模态间和模态内相似关系。实验结果表明,与典型相关分析方法相比,该方法跨媒体检索精度较高。

关键词: 跨媒体检索, 特征选择, 耦合字典学习, 图像正则化, 特征映射

Abstract:

The method of cross-media retrieval mostly maps the original features of two modalities to the common subspace,and performs cross-media retrieval in the subspace,ignoring the selection of discriminant features and the relationship between modalities.Therefore,a new cross-modal retrieval method based on coupled dictionary learning and graph regularization is proposed.A uniform sparse representation is generated for different modalities by associating and jointly updating dictionaries of different modalities.The sparse representations of the different modalities are then projected into the common subspace defined by the class label information to perform cross-modal matching while applying 21 norm terms to the projection matrix to select the correlation and discriminative features of the feature space.On this basis,the regularization term of the graph is used to preserve the inter-modal and intra-modal similar relationship.Experimental results show that compared with the Canonical Correlation Analysis(CCA) method,the method has higher accuracy in cross-media retrieval.

Key words: cross-media retrieval, feature selection, coupled dictionary learning, graph regularization, feature mapping

中图分类号: