Self-recovery Fragile Watermarking For Natural&Text Image Authentication

With the development of digital cameras,smartphones and other imaging equipment,people can easily access digital images.With the development of computer technology and image processing,digital image applications play an indispensable role in daily life in various fields.However,image processing software also facilitates attackers to illegally edit,retouch,and even forge digital images.Therefore,the authenticity and integrity of digital images has become an important research branch in the field of information security.The goal of this thesis is to carry out the research of self-recovery watermark algorithm of digital image with the ability of tampering localization and tampering recovery by using the basic theory of information hiding and cryptography.Self-recovering watermark uses watermark information embedded in image content to locate the area of digital images that have been tampered with,and also can approximate the recovery of image content in the altered area by using watermark information extracted from untampered areas.On the basis of summarizing and analyzing the latest research progress of the existing self-recovery watermark,the thesis combines different application scenarios,takes natural images and text images as the research object,and carries out the research of digital image self-recovery watermark algorithm,and the research results mainly include the following three aspects.At first scheme,we proposed a restorable fragile image watermarking scheme.The main objective of this method is to achieve good tamper detection and recovery performance for the potential tampered natural objects,characters,and numbers in the image.In this work generates the authentication watermark for each 4 × 4 blocks by a hash function and a variable capacity recovery watermark for each 2 × 2 blocks based on DCT by allocating more bits to the textural blocks and fewer bits to the smooth ones to reduce the watermark capacity.The authentication watermark and the recovery one are embedded in the original image by adopting different strategies based on a secret key.The multi-stage neighbor detection strategy is designed to locate the tampered image blocks accurately.The simulation results show that the proposed scheme achieves better recovery quality and invisibility for natural and text images due to a variable-capacity recovery watermark and superior localization.The proposed multi-stage tamper detection performs superiorly under various malicious attacks such as general tampering,collage,content-only,and a hybrid attack.Further,the computationally of the proposed method is less.In the second scheme,we proposed a self-recovery fragile image watermarking scheme based on block truncation coding and quantum chaos map.In the proposed algorithm,the original image is divided into 2×2 blocks.For each block,the block truncation coding is using to generate recovery data and authentication data to reduce the watermark capacity,recovery watermark and authentication watermark are embedded in the spatial domain using the LSB technique.Quantum chaos map is using to generate a recovery watermark embedding location and enlarging the key space to improve the shortcoming of watermarking such as small key space and low security.Least significant methods are used to embed the watermark information to improve the imperceptibility of the watermarked image.The experimental results show that the proposed scheme introduces less distortion into the watermarked images and provides a better quality of the recovered image.Additionally,this approach achieves higher security under different attacks,such as general tampering,collage attack,content-only,and hybrid attack,and also the key space is large enough to resist the more attack.In the third scheme,we proposed a self-recovery fragile watermarking scheme based on block-BTC combined with DCT alterable-coding.The recovery watermark is generated by encoding the block content,the DCT is implemented in each rough block,and the BTC is implemented in each smooth block to generate the codes bits to reduce the recovery watermark capacity.The authentication watermark is generated with a location pixel to improve the performance of tamper detection.The recovery watermark is randomly embedded in the mapping blocks based on a secret key and authentication watermark embedded in the original block.Combination the recovery watermark and authentication watermark is uses to improve the localization accurately.The simulation results show that the proposed scheme achieves higher invisibility and better recovery quality for natural and text images.