Toward high-capacity crypto-domain reversible data hiding with huffman-based lossless image coding.

The rapid development of the Internet of Things has opened up commercial and economic opportunities, but there are also risks and threats to cybersecurity: a significant amount of private data is exposed in an untrusted cyberspace. In view of this issue, the research on reversible data hiding in encrypted image (RDHEI) aims to establish a secure and private communication channel between a sender and a recipient by embedding data into encrypted images. Therefore, this paper proposes a novel RDHEI scheme that employs a Huffman-based lossless image coding technique to achieve a high embedding capacity. Specifically, we compress the original image into a bitstream through the JPEG-like compression and then further encode the difference values between the reconstructed JPEG image and the original image via the Huffman coding. Due to the high compression rate and coding efficiency of the combined strategy, a large embedding space can be created. Our proposed scheme not only ensures a lossless recovery of the original image but also an error-free extraction of the secret messages. Several experiments are conducted to verify that our proposed scheme achieves a state-of-the-art performance compared with the prior art in terms of embedding capacity. [ABSTRACT FROM AUTHOR]