Research On The Technology Of Image Authentication And Its Applications

During the transmission of digital image, the protection of image authenticity and integrality protection continues to be a concern due to the image processing tools have become increasingly powerful that can access and tampered digital images easily. As a kind of efficient and promising technology to achieve secure transmission of digital images, image authentication has attracted increasing attention to address issues of image authenticity verification and integrity protection. There are two typical schemes of image authentication which are digital-signature based schemes and fragile watermarking based schemes. Simultaneously, the technology of image authentication is very extensive that used in any type of image transmission. In this thesis, we focus on the applications of image authentication that are the compressed image and secret image sharing.Firstly, a novel fragile watermark-based image authentication scheme for images that are compressed by absolute moment block truncation coding (AMBTC) was proposed in this thesis. In the proposed scheme, an authentication code is generated by a pseudo random sequence. Subsequently, the authentication code is embedded into both quantization levels and the bit map of each image block for AMBTC-compressed images. Unlike other schemes, it is not necessary to consider various cases and obtain different image quality according to the number of bits of the authentication code in binary form. Experimental results demonstrate that our new scheme provides good image quality for watermarked images. Moreover, the high image quality is not influenced by an increasing authentication code.Secondly, secret image sharing has attracted extensive attentions to store and transfer digital image securely. We propose a novel secure secret image sharing scheme based on polynomial with authentication that the authentication code is concealed during the processes of sharing instead of embedding it into the shadows. Subsequently, we utilized the avoidance strategy to make sure the lossless reconstruction of the secret image without any extra bits and the size of the shadows is reduced to half the size of the secret image. Theoretical analysis and experimental results demonstrate that our new scheme achieves the ability of anti-attack of steganalysis and the accuracy of tamper-localization is acceptable.Thirdly, we proposed an adaptive (K,N)-threshold secret image sharing scheme with authentication based on the method of matrix embedding in (n,k)-Hamming code. The proposed scheme can improve the image quality of shares furtherly that compared to Wu et al.’s scheme. Meanwhile, to make sure the ability of tamper detection and the security of authentication, we also use the HMAC to generate the authentication code according to the information of the pixels of secret images and the maximum length of the authentication code is as long as possible to achieve the efficient verification. Of course, our scheme is not so better when the threshold value is less the 3. We will solve this problem in the follow-up work.Finally, we conclude our thesis and summarize several key problems that still exist in the image authentication and point out the direction of future research.