Journal of Computer Applications ›› 2025, Vol. 45 ›› Issue (2): 616-623.DOI: 10.11772/j.issn.1001-9081.2024030282

• Multimedia computing and computer simulation • Previous Articles    

Image watermarking method combining attention mechanism and multi-scale feature

Tianqi ZHANG, Shuang TAN(), Xiwen SHEN, Juan TANG   

  1. School of Communication and Information Engineering,Chongqing University of Posts and Telecommunications,Chongqing 400065,China
  • Received:2024-03-18 Revised:2024-06-20 Accepted:2024-06-25 Online:2024-10-14 Published:2025-02-10
  • Contact: Shuang TAN
  • About author:ZHANG Tianqi, born in 1971, Ph. D., professor. His research interests include modulation and demodulation of communication signals, blind processing.
    SHEN Xiwen, born in 2000, M. S. candidate. His research interests include speech enhancement, speech signal processing.
    TANG Juan, born in 2000, M. S. candidate. Her research interests include satellite spread spectrum signal capture.
  • Supported by:
    Natural Science Foundation of Chongqing(cstc2021jcyj-msxmX0836)

融合注意力机制和多尺度特征的图像水印方法

张天骐, 谭霜(), 沈夕文, 唐娟   

  1. 重庆邮电大学 通信与信息工程学院,重庆 400065
  • 通讯作者: 谭霜
  • 作者简介:张天骐(1971—),男,四川眉山人,教授,博士,CCF会员,主要研究方向:通信信号的调制解调、盲处理
    沈夕文(2000—),男,安徽滁州人,硕士研究生,主要研究方向:语音增强、语音信号处理
    唐娟(2000—),女,四川德阳人,硕士研究生,主要研究方向:卫星扩频信号捕获。
  • 基金资助:
    重庆市自然科学基金资助项目(cstc2021jcyj-msxmX0836)

Abstract:

Aiming at the problems that the watermarking method based on deep learning does not fully highlight key features of the image and does not utilize the output features of the intermediate convolution layer effectively, to improve the visual quality and the ability to resist noise attacks of the watermarked image, an attention mechanism-based multi-scale feature image watermarking method was proposed. An attention module was designed in the encoder part to focus on important image features, thereby reducing image distortion caused by watermark embedding; a multi-scale feature extraction module was designed in the decoder part to capture different levels of image details. Experimental results show that compared with the deep watermark model HiDDeN(Hiding Data with Deep Networks) on COCO dataset, the proposed method has the generated watermarked image’s Peak Signal-to-Noise Ratio (PSNR) and Structural SIMilarity (SSIM) increased by 11.63% and 1.29% respectively and has the average Bit Error Rate (BER) of watermark extraction for dropout, cropout, crop, Gaussian blur, and JPEG compression reduced by 53.85%. In addition, ablation experimental results confirm that the method adding attention module and multi-scale feature extraction module has better invisibility and robustness.

Key words: image watermarking, attention mechanism, feature extraction, robust watermarking, deep learning, adversarial training

摘要:

针对基于深度学习的水印方法未充分突显图像的关键特征,以及未有效利用中间卷积层输出特征的问题,为提升含水印图像的视觉质量和抵抗噪声攻击的能力,提出一种融合注意力机制和多尺度特征的图像水印方法。在编码器部分,设计注意力模块关注重要图像特征,以减小水印嵌入引起的图像失真;在解码器部分,设计多尺度特征提取模块,以捕获不同层次的图像细节。实验结果表明,在COCO数据集上与深度水印模型HiDDeN(Hiding Data with Deep Networks)相比,所提方法生成的含水印图像的峰值信噪比(PSNR)和结构相似度(SSIM)分别增加了11.63%和1.29%;所提方法针对dropout、cropout、crop、高斯模糊和JPEG压缩的水印提取平均误比特率(BER)降低了53.85%;此外,消融实验结果验证了添加注意力模块和多尺度特征提取模块的方法有更好的不可见性和鲁棒性。

关键词: 图像水印, 注意力机制, 特征提取, 鲁棒水印, 深度学习, 对抗训练

CLC Number: