Enhancing Data Embedding Capacity in Encrypted Images via Adaptive MSB Prediction

With the advancement of cloud storage technology, people have become accustomed to storing various types of information in the cloud. In the context of cloud storage applications, scholars have proposed the concept of Reversible Data Hiding in Encrypted Images (RDHEI). RDHEI can be categorized into two main types: Reserving Room Before Encryption (RRBE) and Vacating Room After Encryption (VRAE). Particularly in VRAE, to preserve the correlation between pixels in an image, the method of Image Block Lossless Compression is used for hiding confidential information. Under this approach, the more similar pixels there are within an Image Block, the greater the capacity for hiding confidential information. However, this also results in a reduction in the number of usable Image Blocks in the image. To address this issue, this study proposes a novel approach of recombining multiple Image Blocks into Image Groups. This method increases the number of similar pixels within Image Blocks while maintaining a high proportion of usable Image Blocks. Experimental results have demonstrated that the method proposed in this study effectively increases the capacity for hiding confidential information in encrypted images. Additionally, it ensures that the encrypted images can be completely restored, highlighting the method's efficacy in balancing the requirements of both high-capacity data hiding and full image recovery.