• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

计算机工程与科学

• 论文 • 上一篇    下一篇

基于类编码的判别特征学习

徐德荣,陈秀宏,田进   

  1. (江南大学数字媒体学院,江苏 无锡 214122)
  • 收稿日期:2016-10-17 修回日期:2016-11-23 出版日期:2018-03-25 发布日期:2018-03-25
  • 基金资助:

    国家自然科学基金(61373055);江苏省2015年度普通高校研究生实践创新项目(SJLX150566)

XU Derong,CHEN Xiuhong,TIAN Jin   

  1. (School of Digital Media,Jiangnan University,Wuxi 214122,China)
  • Received:2016-10-17 Revised:2016-11-23 Online:2018-03-25 Published:2018-03-25

摘要:

经典的自编码模型(BAE、SAE、DAE、CAE)都是基于输出数据对原始数据的重构,提取输入信息的低维度特征,将该特征用于图像分类不一定能够取得很好的判别效果。利用标签信息,提出了堆叠判别自编码模型(SDcAE),该模型将类编码作为隐层神经元约束加入到堆叠自编码器的训练中,使得隐层学习的特征具有更好的判别能力。同时,将类编码作为判别损失加入到Softmax分类器中,提出了类编码分类器(CEC)。由于类间样本特征误差的降低,该分类器可以取得更好的训练效果,从而提高了最终分类的正确率。实验表明,堆叠判别自编码器和类编码分类器在图像分类中是有效可行的。
 

关键词: 类编码, 堆叠判别自编码, 类编码分类器, 图像分类

Abstract:

Based on the output data’s reconstructing the original data, classical auto encoders(BAE、SAE、DAE、CAE) can extract low dimensional features of the input information.Applying these features in image classification may not able to guarantee a good result.In this paper,we use the label information to propose a stacked discriminant auto encoder (SDcAE),which adds the class encoderas the constraint of hidden layer neuron into the training process of the stacked auto encoder. Hence, the features learned by the hidden layer have better discrimination ability. In addition,we propose a class encoding classifier(CEC),which adds the class encoder as the discriminative loss into the Softmax classifier.Due to the decrease of the feature error of the interclass samples, the classifier can achieve better training results, thus improving the final classification accuracy.The experimental results show that the stacked discriminant auto encoder and the class encoding classifier are effective and feasible in image classification.
 

Key words: class encoder, stacked discriminant auto encoder, class encoding classifier, image classification