计算机科学 ›› 2026, Vol. 53 ›› Issue (1): 413-422.doi: 10.11896/jsjkx.241100040
张小敏, 赵军智, 和红杰
ZHANG Xiaomin, ZHAO Junzhi, HE Hongjie
摘要： 屏摄鲁棒的水印算法在版权保护、追踪溯源等领域具有重要意义。现有的抗屏摄鲁棒水印算法大多关注于自然图像,忽视了对文档图像的研究。文档载体本身的冗余信息较少,水印的鲁棒性和不可感知性很难得到平衡。为解决这一问题,提出了一种基于注意力机制的文档图像屏摄鲁棒水印方法。为提升水印的不可感知性,在编码器网络中引入注意力特征融合模块,实现浅层特征和深层特征的自适应聚合。为提高算法的鲁棒提取能力,在解码器网络中设计了自适应通道与空间注意力模块,突出通道和空间维度中与水印信息高度相关的特征。同时,在屏摄噪声模拟过程中设计了摩尔纹失真层,以提高算法抵抗真实摩尔纹干扰的鲁棒性能。实验结果显示,所提方法的平均PSNR为33.4 dB,SSIM为0.988 5,RMSE为5.48,在多种屏摄场景的平均提取准确率可达99.49%。在图像质量和水印鲁棒性方面,均优于现有同类方法。
中图分类号:
| [1]GUAN H,HUANG Y,ZHANG S,et al.A Survey of Screen Shooting Resilient Image Watermarking Algorithms[C]//2023 International Conference on Culture-Oriented Science and Technology.IEEE,2023:12-17. [2]FANG H,ZHANG W,ZHOU H,et al.Screen-shooting resilient watermarking[J].IEEE Transactions on Information Forensics and Security,2018,14(6):1403-1418. [3]FANG H,ZHANG W,MA Z,et al.A camera shooting resilient watermarking scheme for underpainting documents[J].IEEE Transactions on Circuits and Systems for Video Technology,2019,30(11):4075-4089. [4]CHENG Y,JI X,WANG L,et al.{mID}:Tracing screen photos via {Moiré} patterns[C]//30th USENIX Security Symposium.2021:2969-2986. [5]GUGELMANN D,SOMMER D,LENDERS V,et al.Screen watermarking for data theft investigation and attribution[C]//2018 10th International Conference on Cyber Conflict.IEEE,2018:391-408. [6]WENGROWSKI E,DANA K.Light field messaging with deep photographic steganography[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2019:1515-1524. [7]TANCIK M,MILDENHALL B,NG R.Stegastamp:Invisiblehyperlinks in physical photographs[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2020:2117-2126. [8]FANG H,JIA Z,MA Z,et al.Pimog:An effective screen-shooting noise-layer simulation for deep-learning-based watermarking network[C]//Proceedings of the 30th ACM International Conference on Multimedia.2022:2267-2275. [9]GE S,FEI J,XIA Z,et al.A screen-shooting resilient document image watermarking scheme using deep neural network[J].IET Image Processing,2023,17(2):323-336. [10]PENG W P,LIU J B,PING Y,et al.Model protection scheme for fusion of internal and external feature watermarks[J].Journal of Chongqing University of Posts and Telecommunication(Natural Science Edition),2024,36(4):765-774. [11]RONNEBERGER O,FISCHER P,BROX T.U-Net:Convolu-tional networks for biomedical image segmentation[C]//Medical Image Computing and Computer-Assisted Intervention(MICCAI 2015).Springer,2015:234-241. [12]YUAN S,TIMOFTE R,SLABAUGH G,et al.Aim 2019 challenge on image demoireing:Methods and results[C]//2019 IEEE/CVF International Conference on Computer Vision Workshop.IEEE,2019:3534-3545. [13]HE K,ZHANG X,REN S,et al.Deep residual learning forimage recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2016:770-778. [14]CHEN Z,HE Z,LU Z M.DEA-Net:Single image dehazingbased on detail-enhanced convolution and content-guided attention[J].arXiv:2301.04805,2023. [15]DAI Y,GIESEKE F,OEHMCKE S,et al.Attentional featurefusion[C]//Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision.2021:3560-3569. [16]JIA Z,FANG H,ZHANG W.Mbrs:Enhancing robustness of dnn-based watermarking by mini-batch of real and simulated jpeg compression[C]//Proceedings of the 29th ACM International Conference on Multimedia.2021:41-49. [17]HU J,SHEN L,SUN G.Squeeze-and-excitation networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2018:7132-7141. [18]WOO S,PARK J,LEE J Y,et al.Cbam:Convolutional block at-tention module[C]//Proceedings of the European Conference on Computer Vision.2018:3-19. [19]SHIN R,SONG D.Jpeg-resistant adversarial images[C]//NIPS 2017 Workshop on Machine Learning and Computer Security.2017:8. [20]ZHANG R,ISOLA P,EFROS A A,et al.The unreasonable effectiveness of deep features as a perceptual metric[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2018:586-595. [21]GE S,XIA Z,FEI J,et al.A robust document image watermar-king scheme using deep neural network[J].Multimedia Tools and Applications,2023,82(25):38589-38612. [22]KINGMA D P,BA J.Adam:A method for stochastic optimiza-tion[J].arXiv:1412.6980,2014. |
|
||
首页