Research On Data Hiding Of High Dynamic Range Image

High dynamic range(HDR)imaging has received extensive attention and rapid development.However,the HDR image is likely to encounter some challenges,including the copyright protection,content authentication and integrity and so on.This dissertation deeply analyses and excavates the characteristics of storage format and visual perception of HDR images,and then proposes some information hiding algorithms from different perspectives.The concrete research content is as follows:1.For covert communication,a steganography algorithm of HDR image is proposed in this paper.Secret information can be embedded effectively in the appropriate significance bit of decimal floating-point number for each pixel.The algorithm has excellent invisibility,and can not only greatly expanded the embedding capacity but also keep the statistical regularity of the pixel value unchanged,which is superior to the existing steganography algorithm of HDR image.2.For copyright protection,a blind tone mapping-robust watermarking algorithm for HDR images is proposed.First,two important spatial activity concepts are respectively defined to characterize the spatial diversity of the robustness and imperceptibility of tone-mapped images.Then,nonsubsampled contourlet transform and singular value decomposition are successively performed to extract the associated structural information.In addition,hierarchical embedding intensity and hybrid perceptual mask are designed to enhance the imperceptibility and robustness of the HDR image watermarking.Experiment results show that the proposed algorithm is superior to the current state-of-the-art watermarking algorithms.3.For content authentication,a tamper detection and recovery algorithm for HDR image is proposed via self-embedding.The host image is embedded with watermark bits including the reference data for content recovery and the authentication data for tampering location.On the one hand,different compression methods for each channel of HDR image are designed to optimize the storage space and image quality of reference data.On the other hand,we explore the block sampling mechanism and self-embedding strategy to reduce the probability of tampering coincidence problem and the computational complexity of the algorithm.The experimental results show that the proposed algorithm can locate and recover the tamper region of HDR image effectively,with high tamper recovery rate and low computational complexity.4.For sensitive areas,we propose a prediction error expansion based data hiding algorithm which can embed reversible watermark in the HDR image with low distortion.On the one hand,the multiple carriers are generated to improve the embedding capacity.On the other hand,multi-modal predictor and multi-level embedding strategy are adopted adaptively to realize multi-level reversible HDR image information hiding based on prediction error expansion.The experimental results show that the proposed algorithm has good invisibility.Meanwhile,the combination of the multiple carriers,multi-modal predictor and multi-level embedding strategy,further improve the embedding capacity and security of the algorithm.