Journal of Computer Applications ›› 2016, Vol. 36 ›› Issue (3): 713-717.DOI: 10.11772/j.issn.1001-9081.2016.03.713

Previous Articles     Next Articles

Image recognition algorithm based on dual-view discriminant correlation analysis

LI Jin, QIAN Xu   

  1. School of Mechanical Electronic and Information Engineering, China University of Mining and Technology, Beijing, Beijing 100083, China
  • Received:2015-10-27 Revised:2015-12-02 Online:2016-03-10 Published:2016-03-17
  • Supported by:
    This work is partially supported by the National Natural Science Foundation of China (61103171).

基于双重鉴别相关性分析的图像识别算法

李晋, 钱旭   

  1. 中国矿业大学(北京) 机电与信息工程学院, 北京 100083
  • 通讯作者: 李晋
  • 作者简介:李晋(1985-),男,山西大同人,博士研究生,主要研究方向:人工智能、模式识别及机器学习;钱旭(1962-),男,北京人,教授,博士,主要研究方向:数据库、信息融合、软件工程、计算机支持的协同工作。
  • 基金资助:
    国家自然科学基金资助项目(61103171)。

Abstract: Focusing on the issue that multi-view correlation analysis are not effective to exploit the correlation information and neglect latent discriminant information in images, a Dual-View Discriminant Correlation Analysis (DVDCA) approach based on dual view was proposed. Firstly, the supervised within-class correlation variation and between-class correlation variation were designed; secondly, within-class correlation variation was maximized and between-class correlation variation was minimized to extract the discriminant feature; finally, constrained dual-view discriminant correlation model was designed to exploit rich view information of both within-view and between-view. Compared with multi-view linear discriminant analysis, Canonical Correlation Analysis (CCA), Multi-view Discriminant Latent Space (MDLS), Uncorrelated Multi-view Discrimination Dictionary Learning (UMDDL) on the Multi-PIE dataset, the proposed algorithm can achieve recognition rate increase of 1.45-4.73 percentage points; on the MFD dataset, the proposed algorithm can achieve increase of 1.25-5.29 percentage points.

Key words: discriminant analysis, image recognition, multi-view learning, Canonical Correlation Analysis (CCA), information mining, dual-view discriminant analysis

摘要: 针对多视图相关性算法未有效利用视图中相关信息且忽视了潜在的鉴别信息的问题,提出基于同一视图内和不同视图间的双重鉴别相关性分析(DVDCA)算法。首先,设计有监督的类内和类间相关性变量,通过最大化类内相关性变量、最小化类间相关性变量来提取视图中的鉴别特征;其次,考虑在同一视图内和不同视图间均考虑进行鉴别相关特征提取,设计约束形式的双重视图鉴别相关性特征提取模型,以利用丰富的视图信息。在Multi-PIE多角度人脸数据集数据集上与多视图线性鉴别分析、典型相关性分析(CCA)、多视图鉴别隐性空间(MDLS)、不相关多视图鉴别字典学习(UMDDL)四种算法对比实验,DVDCA分类识别率能够提高1.45~4.73个百分点;在MFD多特征手写体数据集上分类识别率能够提高1.25~5.29个百分点。

关键词: 鉴别分析, 图像识别, 多视图学习, 典型相关性分析, 信息挖掘, 双重鉴别分析

CLC Number: