期刊文献+

基于条件深度卷积生成对抗网络的图像识别方法 被引量:143

Image Recognition With Conditional Deep Convolutional Generative Adversarial Networks
下载PDF
导出
摘要 生成对抗网络(Generative adversarial networks,GAN)是目前热门的生成式模型.深度卷积生成对抗网络(Deep convolutional GAN,DCGAN)在传统生成对抗网络的基础上,引入卷积神经网络(Convolutional neural networks,CNN)进行无监督训练;条件生成对抗网络(Conditional GAN,CGAN)在GAN的基础上加上条件扩展为条件模型.结合深度卷积生成对抗网络和条件生成对抗网络的优点,建立条件深度卷积生成对抗网络模型(Conditional-DCGAN,C-DCGAN),利用卷积神经网络强大的特征提取能力,在此基础上加以条件辅助生成样本,将此结构再进行优化改进并用于图像识别中,实验结果表明,该方法能有效提高图像的识别准确率. Generative adversarial network(GAN) is a prevalent generative model. Deep convolutional generative adversarial network(DCGAN), based on traditional generative adversarial networks, introduces convolutional neural networks(CNN) into the training for unsupervised learning to improve the effect of generative networks. Conditional generative adversarial network(CGAN) is a conditional model which adds condition extension into GAN. The generative model of conditional-DCGAN(C-DCGAN) is a combination of DCGAN and CGAN, which integrates the feature extraction of convolutional networks and condition auxiliary generative sample for image recognition. The result of simulation experiments shows that this model can improve the accuracy of image recognition.
作者 唐贤伦 杜一铭 刘雨微 李佳歆 马艺玮 TANG Xian-Lun;DU Yi-Ming;LIU Yu-Wei;LI Jia-Xin;MA Yi-Wei(College of Computer Science and Technology, Chongqing University of Posts and Telecommunications, Chongqing 400065;College of Automation, Chongqing University of Posts and Telecommunications, Chongqing 400065)
出处 《自动化学报》 EI CSCD 北大核心 2018年第5期855-864,共10页 Acta Automatica Sinica
基金 国家自然科学基金(61673079 61703068) 重庆市基础科学与前沿技术研究项目(cstc2016jcyj A1919)资助~~
关键词 生成对抗网络 卷积神经网络 条件模型 特征提取 图像识别 Generative adversarial network (GAN) convolutional neural networks (CNN) conditional models featureextraction image recognition
  • 相关文献

参考文献4

二级参考文献170

  • 1王飞跃.平行系统方法与复杂系统的管理和控制[J].控制与决策,2004,19(5):485-489. 被引量:333
  • 2王飞跃.计算实验方法与复杂系统行为分析和决策评估[J].系统仿真学报,2004,16(5):893-897. 被引量:147
  • 3钱跃良,林守勋,刘群,刘洋,刘宏,谢萦.863计划中文信息处理与智能人机接口基础数据库的设计和实现[J].高技术通讯,2005,15(1):107-110. 被引量:4
  • 4王飞跃.关于复杂系统的建模、分析、控制和管理[J].复杂系统与复杂性科学,2006,3(2):26-34. 被引量:64
  • 5Settles B. Active Learning Literature Survey, Computer Science Technical Report 1648, University of Wisconsin- Madison, USA, 2009. 3-4.
  • 6Dasgupta S. Coarse sample complexity bounds for active learning. Advances in Neural Information Processing Sys- tems. Cambridge: The MIT Press, 2006. 235-242.
  • 7Tong S, Chang E. Support vector machine active learning for image retrieval. In: Proceedings of the 9th ACM Inter- national Conference on Multimedia. New York, USA: ACM, 2001. 107-118.
  • 8Tong S, Koller D. Support vector machine active learning with applications to text classification. The Journal of Ma- chine Learning Research, 2002, 2:45-66.
  • 9Seung H S, Opper M, Sompolinsky H. Query by commit- tee. In: Proceedings of the 5th Annual Workshop on Com- putational Learning Theory. New York, USA: ACM, 1992. 287-294.
  • 10Dagan I, Engelson S P. Committee-based sampling for train- ing probabilistic classifiers. In: Proceedings of the 12th International Conference on Machine Learning. California, USA: Morgan Kaufmann, 1995. 150-157.

共引文献920

同被引文献961

引证文献143

二级引证文献1364

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部