Dual Parted Pairwise PEE with Adaptable IPVO for High Fidelity Images

Abstract

In this study, we explore the practical application of Reversible Data Hiding (RDH) in digital images using Adaptive Integer Pixel-value Ordering and Verification (Adaptive IPVO) and Two-Segment Pairwise Pixel Embedding and Extraction (Two-Segment Pairwise PEE). Our project employs Python and its robust libraries, namely NumPy and Pillow, to manipulate image and data. The system successfully embeds binary data into an image, then extracts and restores the original image, demonstrating RDH's reversibility. We also evaluate the performance using key metrics such as payload capacity, embedding efficiency, and Peak Signal-to-Noise Ratio (PSNR). Future work aims at improving robustness, error handling, and user-friendliness, facilitating the use of RDH for broader audiences.