A Novel Reversible Data Hiding Scheme in Compressible Encrypted Grayscale-Based Images

In recent years, the proliferation of cloud services, particularly social media, has spurred research on encryption-then-compression (EtC) systems for secure and efficient multi-media transmission. EtC systems compress encrypted images or provide compression-resilient encryption. To address the need for embedding authentication, copyright data, or tags to images in cloud services, this paper introduces a novel data hiding scheme for the EtC system. The EtC system which this paper focuses on converts an RGB color image to the YCbCr color image, followed by horizontally joining three color component images to form a grayscale image, and encrypts this grayscale image in non-overlapping blocks using block permutation, block rotation/inversion, and brightness inversion. The proposed data hiding scheme conceals data in unencrypted images and can extract it from encrypted images by manipulating diagonal pixel values within blocks. Sorting keys, generated from intra-block pixel value characteristics, ensure correct data sequence recovery. Moreover, the proposed scheme is reversible, allowing for the removal of embedded information, and uses different pixels within the same block for dual embeddings to increase payload. Compared to conventional schemes for the EtC system, the proposed scheme leverages all three visual encryption processes and considers the RGB-to-YCbCr conversion. Performance evaluation results confirm the effectiveness of the proposed scheme.