计算机与现代化 ›› 2024, Vol. 0 ›› Issue (04): 38-42.doi: 10.3969/j.issn.1006-2475.2024.04.007

• 图像处理 • 上一篇    下一篇

基于I-ConvNeXt的GAN生成人脸图像鉴别方法

  



  1. (1.华中师范大学物理科学与技术学院,湖北 武汉 430079; 2.中船重工武汉船舶通信研究所,湖北 武汉 430079)
  • 出版日期:2024-04-30 发布日期:2024-05-13
  • 作者简介:肖梦思(1998—),女,湖北黄石人,硕士研究生,研究方向:数字媒体取证,E-mail: xiaomengsi@mails.ccnu.edu.cn; 吴建斌(1972—),男,湖北黄梅人,副教授,博士,研究方向:信息隐藏,E-mail: wujianbin@mail.ccnu.edu.cn; 涂雅蒙(1992—),女,湖北荆州人,博士研究生,研究方向;数字媒体取证,E-mail: tymeng@mails.ccnu.edu.cn; 袁林锋(1972—),男,湖北黄梅人,高级工程师,博士,研究方向:通信系统,E-mail: yuanlf@163.com。
  • 基金资助:
    国家自然科学基金资助项目(U1736121); 中央高校基本科研业务费专项资金资助项目(CCNU22JC024)

GAN-generated Fake Images Recognition Based on Improved ConvNeXt


  1. (1. College of Physics Science and Technology, Central China Normal University, Wuhan 430079, China;
    2. Wuhan Maritime Communication Research Institute, Wuhan 430079, China)
  • Online:2024-04-30 Published:2024-05-13

摘要:
摘要:为鉴别社交网络中人脸图像的真假,在ConvNeXt基础上提出一种针对生成式对抗网络(Generative Adversarial Networks, GAN)生成人脸图像的鉴别方法。该方法以ConvNeXt网络结构为主体,利用人脸图像的颜色特征和空间纹理特征,采用多颜色空间多通道组合输入(Multichannel Input, MCI),扩大网络的学习范围;同时引入通道注意力机制和空间注意力机制来凸显真假人脸图像在颜色分量和纹理特征上的差异,进而实现生成人脸图像和真实人脸图像的检测与识别。实验结果表明,使用改进后的ConvNeXt (Improved ConvNeXt, I-ConvNeXt)网络结构对GAN生成人脸图像的识别准确率达到了99.405%,与原ConvNeXt算法相比,平均准确率提高了1.455个百分点。该结果验证了所提方案的可行性、合理性。

关键词: 关键词:生成式对抗网络, 注意力机制, 颜色特征, 生成人脸, 多通道输入

Abstract:
Abstract: In order to distinguish the authenticity of face images in social networks, a recognition method based on ConvNeXt for face image generated by Generative adversarial networks (GAN) is proposed. The ConvNeXt network structure is used as the main body, using the color features and spatial texture features of the face image, and multi-channel combination input (Multichannel Input, MCI) with multi-color space is used to expand the learning range of the network, while channel attention mechanism and spatial attention mechanism are introduced to highlight the differences between real and fake face images in color components and spatial features, and then the detection and recognition of fake face images are achieved. The experimental results show that the recognition accuracy of face images generated by GAN with improved ConvNeXt (I-ConvNeXt) network structure reaches 99.405%, with an average accuracy improvement of 1.455 percentage points compared with the original ConvNeXt algorithm. The results validate the feasibility and reasonableness of the proposed scheme.

Key words: Key words: generative adversarial network, attention mechanism, color features, generated face image, multi-channel input

中图分类号: