An Information Security Method Based on Optimized High-Fidelity Reversible Data Hiding

With the rapid development of the Industrial Internet of Things, the social and commercial value of digital information is greatly increased, and the sensitive information of groups and individuals faces more and more problems. Due to providing accurate prediction errors, pixel value ordering (PVO)-based reversible data hiding is an active research topic. In this article, a new PVO-based embedding method is proposed first, in which the adaptive data embedding is adopted according to the difference between the maximum/minimum three pixels. Second, a pixel collecting and sorting model is designed to increase expandable prediction errors. By converting a shifting prediction error into an expandable one, the proposed PVO-based embedding method creates conditions for data embedding. In this way, the space location of pixels in the block can be better utilized to achieve improved capacity-distortion performance. Finally, an adaptive embedding is designed based on the content of the target block and its neighboring blocks. By measuring the complexity of the target pixel block, smoothing blocks are preferentially used and the appropriate embedding is determined by the complexity. The experimental results show that the proposed method outperforms a series of PVO-based multipass embedding methods. The proposed method is expected to address some security issues in the industrial and privacy fields, such as enhancing confidentiality and security during the transmission of sensitive information.