首页 | 本学科首页   官方微博 | 高级检索  
     

An Authentication Technique Based on Classification
引用本文:李钢,杨杰. An Authentication Technique Based on Classification[J]. 东华大学学报(英文版), 2004, 21(3): 103-106
作者姓名:李钢  杨杰
作者单位:Inst.OfImageProcessing&PatternRecognition,ShanghaiJiaotongUniv.,Shanghai200030,China
摘    要:
We present a novel watermarking approach based on classification for authentication,in which a watermark is embedded into the host image.When the marked image is modified,the extracted watermark is also different to the original watermark,and different kinds of modification lead to different extracted watermarks,In this paper,different kinds of modification are considered sa classes,and we used classification algorithm to recognize the modifications with high probability.Simulation results show that the proposed method is potential and effective.

关 键 词:鉴定技术 水印 图像识别 数据隐藏

An Authentication Technique Based on Classification
LI Gang YANG JieInst. Of Image Processing , Pattern Recognition,Shanghai Jiaotong Univ.,Shanghai ,China. An Authentication Technique Based on Classification[J]. Journal of Donghua University, 2004, 21(3): 103-106
Authors:LI Gang YANG JieInst. Of Image Processing & Pattern Recognition  Shanghai Jiaotong Univ.  Shanghai   China
Affiliation:Inst. Of Image Processing & Pattern Recognition, Shanghai Jiaotong Univ., Shanghai 200030, China
Abstract:
We present a novel watermarking approach based on classification for authentication, in which a watermark is embedded into the host image. When the marked image is modified, the extracted watermark is also different to the original watermark, and different kinds of modification lead to different extracted watermarks. In this paper, different kinds of modification are considered as classes, and we used classification algorithm to recognize the modifications with high probability. Simulation results show that the proposed method is potential and effective.
Keywords:Authentication   data hiding   fragile watermark   Classification
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号