| With the progress of information technology,a platform for the rapid development of digital multimedia has been provided.High expansion of Internet technology and storage provides great convenience for the storage,duplication and transmission of multimedia data such as images and videos.At the same time,it offers the possibility of tampering and the theft of digital information.The security of digital information is being threatened.Digital watermarking technology embeds the invisible mark into the original carrier and protects the security of the original carrier effectively.Digital watermarking technology is widely used in image recognition,image retrieval,fingerprint detection and other fields.Digital watermarking technology has made great achievements in the airspace domain,transform domain and compressed domain.Particularly,the appearance of the Just Noticeable Distortion(JND)model has promoted the development of digital watermarking technology.Spread Transform Dither Modulation(STDM)watermarking algorithm,robust JND model based in quantization watermarking and other classical schemes have been produced.In the actual study,we found that the JND model does not consider the visual characteristics and could not show human eye characteristics well.As a result,the digital watermarking system cannot be balanced between robustness and imperceptibility,and the watermarking system cannot meet the requirements of actual application.Based on the reasons above,this paper innovates the JND model and combines the STDM watermarking algorithm to propose a better watermarking scheme.The main achievements of this article are as follows:1.In the search of watermark algorithms,the role of visual characteristics cannot be ignored.The results of recent research showed that the edge is themost important feature which is carrying most of the semantic information in an image.Being motivated by that we provide a novel STDM watermarking scheme using semantic information based JND model in this paper.A new semantic information based in JND model is employed as a product of the edge information.Improved STDM watermarking using semantic information-based JND model is proposed.In this way,the watermarking system can achieve a better balance between robustness and fidelity.2.Recently,the VS(Visual Saliency)model has attracted tremendous attention.As a consequence of evolution,most vertebrates,including humans,have a remarkable ability to automatically pay more attention to salient regions of the visual scene.As we all know,perceptual watermarking should make use of the results from human visual system(HVS)achievements.JND gives us a way to model the HVS accurately.VS is another very important aspect which is affecting human perception.Intuitively,VS and JND are intrinsically related because both of them are depended on how the HVS perceives an image,and suprathreshold distortions can be a strong attractor of visual attention.In this paper,we present a novel STDM watermarking scheme based on an effective visual saliency-based JND model.Combined with the comparison masking factor,brightness masking factor and visual salience features,the visual redundancy vector on the watermark embedder and detector is calculated.Experimental results show that the proposed watermarking scheme has a superior performance compared with existing schemes. |
