Font Size: a A A

The Research Of Geometrically Robust Digital Image Watermarking Based On The Sift Feature Point And Region Of Significance

Posted on:2013-10-18Degree:MasterType:Thesis
Country:ChinaCandidate:H Q ZhaoFull Text:PDF
GTID:2248330374979233Subject:Computer application technology
Abstract/Summary:PDF Full Text Request
With the rapid development of network and digital technology, the multimediainformation is becoming more and more convenient for the legitimate users. And atthe same time there is more opportunity for the illegal users to take advantage of,which results in the growing social problems, such as information security andcopyright protection. So it is hoped that there will be some effective technologymethods to solve these problems. To cater for the social demands, the digitalwatermarking technology, as an important branch of information-hiding technology, isdeveloping to one of the hotpots.The digital watermarking, with characteristics of hidden security and robustness,tend to be more capable to protect the copyright of digital information. The robustnessof digital watermarking plays a decisive role to protect copyright. So it is required thatthe digital watermarking is with good stability to against variety of attacks. Geometricattack is kind of powerful attack but with simple algorithm. So far, there is fewmethod that can cope with this attack. As a result, the algorithm of anti-geometricattack is a hot issue currently.This paper briefly presents two points from the view of enhancing the robustnessof the watermark according to the watermark pretreatment technology and the secondgeneration digital watermarking technology. Firstly, this paper briefly describes somecommon pretreatment methods, mainly introducing the zigzag scrambling technology.For the characteristics to anti-spin attack of the original zigzag scrambling technology,this paper presents the improved program, which has better robustness to anti-spinattack and also keeps the advantage of the original one. Experiments show that theimproved technology is more stable than geometric attacks and does not reduce the effect of scrambling, without significantly increasing the time complexity of theoriginal algorithm. Secondly, according to the second generation digital watermarkingtechnique based on image features of stable, embedding the watermark into multiplelocal feature region characteristic of the image, looking for image stabilization featurepoint area, this paper briefly introduces the image SIFT feature points and significantregional extraction, and then screening out SIFT feature point, which is the relativelystable feature points as the image information mainly concentrates in the obviousregional characteristics. Taking regions in which these feature points concentrates asthe watermark embedding area, we can realize the digital image watermarkembedding. Experimental results show that this algorithm can effectively resistgeometric attacks and conventional signal attack.
Keywords/Search Tags:Digital watermark, Zigzag scrambling, Geometric attack, Featurepoints, Region of significance
PDF Full Text Request
Related items