Research On Adaptive Image Information Hiding Scheme For Robustness And High Quality Recovery

With the rapid development of Internet technology and image processing technology,it has brought great convenience to the transmission and editing of digital images.This has also led to more frequent violations of intellectual property rights and copyright disputes,which has brought substantial loss of interests to people.Therefore,it is particularly important to protect the copyright of digital images and ensure legal use.Digital watermarking technology is an effective information hiding method,which can embed secret information into the host image to protect copyright information and track tampering.However,existing digital watermarking schemes lack certain integrity.For example,under the premise of ensuring a certain embedding capacity,the ability to resist attacks is limited and the quality of encrypted images needs to be improved,it is difficult to maintain the accuracy of tampering detection under different attack situations,and content recovery cannot be achieved when the tampering rate is high,etc.From these perspectives,the main research work of this dissertation is as follows:1)Aiming at the balance between the invisibility and robustness of traditional watermarking methods,a robust digital watermarking algorithm based on feature extraction is proposed.The core idea is to extract the feature points based on the digital image feature analysis algorithm,embed the watermark into the feature points,and realize the embedding and extraction of the digital watermark in the transform domain by transforming the discrete wavelet transform.The experimental data analysis algorithm is used to realize the invisibility of embedding and extraction,and different wavelet bases and wavelet coefficients are used to embed and extract watermark information to test the robustness of the watermark algorithm.The experimental results show that the scheme guarantees the image quality under the condition of strong robustness,and when directly extracting the encrypted copyright information,the normalized correlation coefficient is 1,the quality of the watermarked image is about 55 d B,and the average structure similarity coefficient is The mean is above 0.99.2)For the tampering coincidence problem caused by the single mapping block embedding method,and in the case of attacks,it is difficult to improve the quality of the watermarked image and restore the quality of the image.A high-quality recovery self-embedded watermarking scheme is proposed.The restoration watermark and the authentication watermark containing the image’s information are embedded in the transformation domain and the space domain of the image respectively,it meets the requirements of the image’s robustness against attacks and the requirements of full-blind detection of tampered areas;at the same time,using the embedded watermark,according to the detailed A four-level tampering detection model is designed based on the optimized strategy,and the image content is restored hierarchically according to the restoration accuracy.The experimental results show that the peak signal-tonoise ratio of the dense image has reached 44 d B,and it performs well in collage and column tampering attacks.The peak signal-to-noise ratio of the image is generally 43 d B.When the tampered area is as high as 85%,the quality of the restored image is still higher than 32 d B.In this dissertation,starting from the consistency of information embedding and extraction,the dense image is optimized to improve its overall performance in terms of robustness and invisibility.In addition,this dissertation also studies an algorithm based on self-embedded digital watermarking technology,which can perform tamper detection and recovery,thereby protecting the authenticity and integrity of images.