中国邮电高校学报(英文) ›› 2023, Vol. 30 ›› Issue (5): 1-10.doi: 10.19682/j.cnki.1005-8885.2023.0010

所属专题: Special Topic on Digital Human

• Special Topic : Digital Human •    下一篇

Joint global constraint and Fisher discrimination based multi-layer dictionary learning for image classification

彭宏1,刘耀宗2   

  1. 1. 文化和旅游部民族民间文艺发展中心
    2. 北京邮电大学
  • 收稿日期:2023-07-19 修回日期:2023-09-03 出版日期:2023-10-31 发布日期:2023-10-30
  • 通讯作者: 彭宏 E-mail:466985365@qq.com
  • 基金资助:
    the National Key Research and Development Project (2021YFF0901701).

Joint global constraint and Fisher discrimination based multi-layer dictionary learning for image classification

Hong PENG1,yaozong liu2   

  1. 1. Center for Ethnic and National Folk Literature and Art Development, Ministry of Culture and Tourism of the People's Republic of China, Beijing 100007, China  
    2. School of Computer Science, Beijing University of Posts and Telecommunications, Beijing 100876, China
  • Received:2023-07-19 Revised:2023-09-03 Online:2023-10-31 Published:2023-10-30
  • Contact: Hong PENG E-mail:466985365@qq.com
  • Supported by:
    the National Key Research and Development Project (2021YFF0901701).

摘要:

    A multi-layer dictionary learning algorithm that joints global constraints and Fisher discrimination (JGCFD-MDL) for image classification tasks was proposed. The algorithm reveals the manifold structure of the data by learning the global constraint dictionary and introduces the Fisher discriminative constraint dictionary to minimize the intra-class dispersion of samples and increase the inter-class dispersion. To further quantify the abstract features that characterize the data, a multi-layer dictionary learning framework is constructed to obtain high-level complex semantic structures and improve image classification performance. Finally, the algorithm is verified on the multi-label dataset of court costumes in the Ming Dynasty and Qing Dynasty, and better performance is obtained. Experiments show that compared with the local similarity algorithm, the average precision is improved by 3.34% . Compared with the single-layer dictionary learning algorithm, the one-error is improved by 1.00% , and the average precision is improved by 0.54% . Experiments also show that it has better performance on general datasets.

关键词: global similarity, Fisher discrimination, joint local-constraint and Fisher discrimination based dictionary learning (JLCFDDL), joint global constraint and Fisher discrimination based multi-layer dictionary learning, image classification

Abstract:

    A multi-layer dictionary learning algorithm that joints global constraints and Fisher discrimination (JGCFD-MDL) for image classification tasks was proposed. The algorithm reveals the manifold structure of the data by learning the global constraint dictionary and introduces the Fisher discriminative constraint dictionary to minimize the intra-class dispersion of samples and increase the inter-class dispersion. To further quantify the abstract features that characterize the data, a multi-layer dictionary learning framework is constructed to obtain high-level complex semantic structures and improve image classification performance. Finally, the algorithm is verified on the multi-label dataset of court costumes in the Ming Dynasty and Qing Dynasty, and better performance is obtained. Experiments show that compared with the local similarity algorithm, the average precision is improved by 3.34% . Compared with the single-layer dictionary learning algorithm, the one-error is improved by 1.00% , and the average precision is improved by 0.54% . Experiments also show that it has better performance on general datasets.

Key words:

global similarity, Fisher discrimination, joint local-constraint and Fisher discrimination based dictionary learning (JLCFDDL), joint global constraint and Fisher discrimination based multi-layer dictionary learning, image classification