Robust Digital Watermarking Algorithm And System Design For Macrophotography Images

With the popularization of mobile internet and the rapid development of multimedia technology,it is more and more convenient for mobile portable devices to collect images.Nowadays,mobile devices can meet the needs of different shooting scenarios such as macro and ultra-wide angles,greatly expanding the usage of smartphone photography.Macrophotography images are loved by people for their rich visual details,which results in their increasing frequency of appearance on the internet.While enjoying the use of macrophotography images to share their own life and emotional fun,users also face infringement behaviors such as illegal copying and tampering of images digital information.These infringement behaviors have greatly damaged the interests of copyright owners.Therefore,the problem of macrophotography image copyright protection needs to be solved urgently.Digital watermarking algorithm can well protect image copyright by embedding copyright information in multimedia.Robustness and invisibility are two important indicators of digital watermarking technology.The trade-off between these two factors is a significant challenge in the design of watermarking algorithms.In order to further balance the performance of digital watermarking algorithms,the Just Noticeable Distortions(JND)model has been studied by more and more researchers,which effectively improves the visual quality of digital watermarking algorithms.However,most JND model watermarking algorithms do not consider the visual characteristics of macrophotography images,which consist of both defocused and focused regions,resulting in overestimation or underestimation of visual masking factors.To address this issue,a JND model that incorporates specific masking factors for macrophotography images can be designed and applied to quantization watermarking algorithms.This approach can effectively improve the invisibility and robustness of macrophotography image watermarking algorithms.With the rapid development of mobile internet,the usage rate of mobile devices has surpassed that of PC devices,which leading to an increasing demand for digital watermarking systems on mobile devices.Consequently,it is of great use value to develop the digital watermarking system of the Android and the iOS devices.This thesis focuses on studying a robust digital watermarking algorithm for macrophotography images.By analyzing the impact of sharpness,focus value,texture,direction,and color feature of macrophotography images s on the masking factor,the JND model is enhanced to guide the watermark quantization embedding,resulting in a well-balanced trade-off between robustness and invisibility for macrophotography image watermarking algorithms.Furthermore,robust digital watermarking systems for macrophotography images were designed for Android and iOS mobile devices.The main contents are as follows:1.A Robust Watermarking Algorithm for Macrophotography ImagesThis thesis presents an improved JND model for macrophotography images,which achieves a better balance performance of the macrophotography image watermarking algorithm.Specifically,this thesis introduces a blur masking factor to account for the focusing and sharpness characteristics of macrophotography images.For the texture and direction features in the macrophotography image,an exponential function model is used to simulate the contrast masking effect of the two features,and a contrast masking effect guided by texture and direction metrics is proposed.Additionally,the visual masking effect of color difference in image Cb and Cr channels is considered,and a color complexity masking factor is proposed to design a new JND model.The proposed JND model is applied to the quantization watermarking algorithm.Experiments prove that the JND model designed in this paper has good applicability in macrophotography images and ordinary images,and can well balance the invisibility and robustness of watermarking algorithms.2.Design and Implementation of Robust Watermarking System for Mobile Devices.This thesis designs and implements a robust digital watermarking system for Android and iOS devices.The system comprehensively considers the function and performance requirements of the watermarking system,designs multiple system modules including image acquisition,image preprocessing,watermark preprocessing,watermark embedding extraction,image storage,etc.And the performance tests of the system’s running time test,invisibility test,and anti-social network attack were carried out.The results demonstrate that the digital watermarking systems on Android and iOS are effective in providing image copyright protection while also ensuring a positive user experience.