基于特征重标定生成对抗网络的图像分类算法  

Image classification algorithm based on feature recalibration GAN

作　　者：姜代红[1] 张三友 刘其开 Jiang Daihong;Zhang Sanyou;Liu Qikai(School of Information&Electronic Engineering,Xuzhou Institute of Technology,Xuzhou Jiangsu 221008,China;Dept.of Science&Technology,Suzhou Wujiang District Public Security Bureau,Suzhou Jiangsu 215200,China;School of Information&Electrical Engine-ering,China University of Mining&Technology,Xuzhou Jiangsu 221008,China)

机构地区：[1]徐州工程学院信电工程学院,江苏徐州221008 [2]苏州市吴江区公安局科技信息化大队,江苏苏州215200 [3]中国矿业大学信息与控制工程学院,江苏徐州221008

出　　处：《计算机应用研究》2020年第3期932-935,共4页Application Research of Computers

基　　金：国家自然科学基金资助项目(61379100);江苏省高等学校自然科学研究重大项目(18KJA520012);徐州市科技计划基金资助项目(KC19197)。

摘　　要：针对传统鉴别器的损失策略和结构难以提取到更抽象以及任务相关的鲁棒性特征,从而导致半监督图像分类表现不足的问题,提出了基于特征重标定的生成对抗网络。为了学习到任务相关的特征,在现有半监督GAN的基础上,为鉴别器引入模型在不同状态下的无监督均方差损失正则项,对训练样本中两个分支的同一输入对应得到的不同输出进行参数惩罚,从而指导特征重标定的优化方向。此外,在鉴别器中加入压缩激活模块来优化传统鉴别器的卷积池化结构,该模块自动学习每一个特征通道的重要程度,能够提取任务相关特征并抑制无关特征,实现特征的重标定功能,从而提高半监督图像分类的表现。In view of the loss strategy and structure of traditional discriminator is difficult to extract more abstract and task related robustness features,which leads to the shortage of semi supervised image classification,this paper proposed a generation countermeasure network based on feature recalibration. In order to learn the related features of the task,on the basis of the existing semi supervised GAN,it introduced the unsupervised canonical loss regular term of the model under different states to the discriminator,and the different output of the same input corresponding to the two branches of the training sample was punished to guide the optimization direction of the feature recalibration. In addition,it added the compression activation module to the discriminator to optimize the convolution pool structure of the traditional discriminator. The module automatically learned the importance of each characteristic channel,and could extract the features related to the task to suppress the unrelated features of the task,and realized the recalibration function of the feature,thus improving the performance of the semi-supervised image classification.

关 键 词：生成对抗网络 图像分类 特征重标定 深度学习 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]