期刊文献+

基于Pix2Pix网络的水印去除

Watermark Removal Based on Pix2pix Network
下载PDF
导出
摘要 水印是一种保护版权的有效方法,广泛应用于计算机图像处理,比如图像水印检测、水印去除和反去除等方面.日常从互联网下载的图像,有些不美观且有版权,为了避免带有水印保护的图像版权,一般就丢弃不用或者通过PS方法去掉水印继续使用,后者会造成一些不必要的麻烦.本文研究了Pix2Pix网络模型,利用HSV颜色空间技术去除图片印章,对需要检测的图像查找其水印(印章)等,通过图像二值化、椭圆拟合、图像腐蚀模型和图像轮廓等方法进行检测,明确印章水印位置,实现水印的自动检测,最终利用Pix2Pix技术实现图像水印自动去除.实验结果证明,Pix2Pix网络可以自动去除印章. Watermarking is an effective method to protect copyright,which is widely used in a large number of computer image processing,such as image watermark detection,watermark removal and anti-removal applications.Watermark highlighting is particularly important.Some images downloaded from the Internet are ugly but also have copyright.To avoid the copyright of images with watermark protection,it is generally discarded or removed by PS method to continue to use,which will cause some unnecessary trouble later.In this paper,the Pix2Pix network model is studied.The HSV color space technology is used to remove the image seal,and the watermark(seal)is found for the image to be detected.Through the image binarization,ellipse fitting method,image corrosion model and image contour detection,the position of the seal watermark is clarified,and the automatic detection of the watermark is realized.Finally,the Pix2Pix technology is used to realize the automatic removal of image watermark.Experimental results show that Pix2Pix network can automatically remove the seals.
作者 徐成俊 陈怀圆 XU Cheng-jun;CHEN Huai-yuan(School of Digital Media,Lanzhou University of Arts and Science,Lanzhou 730010,China;School of Electronic and Information Engineering,Lanzhou Jiaotong University,Lanzhou 730070,China)
出处 《兰州文理学院学报(自然科学版)》 2022年第6期64-68,共5页 Journal of Lanzhou University of Arts and Science(Natural Sciences)
基金 甘肃省自然科学基金(22JR5RA217)。
关键词 Pix2pix网络 二值化 空间HSV方法 水印 pix2pix network binarization space HSV method watermark
  • 相关文献

参考文献7

二级参考文献50

  • 1赵永涛,李志敏,王洪建,陈志云,王林.印章识别中的图像预处理研究[J].仪器仪表学报,2004,25(z3):401-403. 被引量:8
  • 2程玉柱,孙星明,黄华军.一种新的基于混沌映射的文本零水印算法[J].计算机应用,2005,25(12):2753-2754. 被引量:15
  • 3卜飞宇,刘长松.一种实用的彩色银行票据二值化方法[J].计算机时代,2007(2):34-37. 被引量:1
  • 4潘梅森,张奋.一种新的图像二值化方法[J].湖南文理学院学报(自然科学版),2007,19(1):72-74. 被引量:3
  • 5Daugman J G.How iris recognition works[J].IEEE Transactions on Circuits System Video Technolody,2004,14( 1 ) :21-30.
  • 6He X F,Shi P F.A new segmentation approach for iris recognition based on hand-held capture device[J].Pattem Recognition,2007,40: 1326-1333.
  • 7Zhu D J,Moore S T,Raphan T.Robust pupil center detection using a curvature algorithm[J].Computer Methods and Programs in Biomedicine, 1999,59: 145-157.
  • 8Zheng Z L,Yang J,Yang L M.A robust method for eye features extraction on color image[J].Pattern Recognition Letters,2005,26: 2252-2261.
  • 9Nair P S,Saunders Jr A T.Hough transform based ellipse detection algorithm[J].Pattern Recognition Letters, 1996,17:777-784.
  • 10Xu L,Oja E,Kuhanen P.A new curve detection method:Randomized Hough Transform(RHT)[J].Pattern Recognition Letters,1990, 11:331-338.

共引文献60

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