基于神经网络的图像可见水印移除模型  

Image Visible Watermark Removal Model Based on Neural Network

作　　者：李祥 LI Xiang(School of Microelectronics,Hefei University of Technology,Hefei Anhui 230601,China)

机构地区：[1]合肥工业大学微电子学院,安徽合肥230601

出　　处：《信息与电脑》2022年第9期77-79,共3页Information & Computer

摘　　要：本文在神经网络的基础上,提出一种基于图像可见水印定位和分割的水印移除模型。模型包含水印提取模块和水印移除模块,分别负责分割水印和移除水印。实验结果表明,本文所提出的模型,在图像可见水印的定位和分割实验中实现了0.99的结构相似度和30的峰值信噪比。在最终的移除水印结果图上,实现了0.90的结构相似度和29的峰值信噪比。A watermark removal model based on neural network is proposed to locate and segment the image visible watermark.The model contains the watermark extraction module and the watermark removal module. Experimental results reveal that the proposed model achieves the SSIM of 0.99 and the PSNR of 30 on the location and segmentation of visible watermark. For the final result, the SSIM of 0.90 and the PSNR of 29 are achieved.

关 键 词：神经网络 生成对抗网络 可见水印分割 可见水印移除 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]