Reversible data hiding is a data-hiding technique which has the ability to recover the original version from stego-images after the secret information is extracted. In this paper, we propose a novel reversible data-hiding scheme based on an enhanced reduced difference-expansion technique. In the proposed scheme, the original image is divided into non-overlapping quad-blocks for embedding data. Then, to enhance the security, the secret bits are encrypted based on the encryption key and a symmetry-based strategy. To improve embedding capacity further, two adjacent encrypted bits are converted into a corresponding decimal digit. Difference expansion (DE) technique is applied to embed a decimal form instead of a binary version. Moreover, to maintain the good image quality, the enhanced reduced difference-expansion technique is used to reduce the original difference values so that it is suitable for decimal embedding. The experimental results demonstrated that the proposed scheme has achieved better performance in comparison with previous solutions.

Digital watermarking is a technique that is widely used to protect rightful ownership of digital images. In this paper, we proposed a hybrid watermarking scheme for stereo images copyright protection. According to the property of stereo images, similar block pairs are located by using a certain key. Then, each block pair is transformed into DCT domain and DCT coefficients are extracted from the anti-diagonal of such block pair to form matrix A. Watermark image is embedded into singular values by applying SVD transformation on the matrix A. The experimental results showed that the proposed scheme can resist to different types of image attacks, such as salt and pepper, noise, and gaussian (density up to 5%). In addition the bit correlation ratio of extracted watermarked images is obtained up to 95%.

This study examines improvements in stegano-graphic models by redesigning three main components: the encoder, decoder, and evaluator, each optimized for better performance. The encoder is designed to embed data effectively with minimal quality, using popular activation function like CeLU. The decoder focuses on accurately retrieving embedded data, essential when images are processed under various conditions. The evaluator uses new functions to accurately check how well data is recovered, promoting continuous improvements. Experimental tests conducted on the Div2K and COCO datasets revealed varied results. The Basic model managed to embed a lot of data (BPP of 4.160), but this reduced image quality significantly, as shown by a PSNR of 30.026 and SSIM of 0.824. In contrast, the Residual model preserved high image quality, with a PSNR of 37.113 and SSIM over 0.9, but embedded less data (BPP of 4.127). The Dense model struck a good balance, embedding a fair amount of data (BPP of 4.512) while maintaining decent image quality (PSNR of 37.089 and SSIM of 0.907). These outcomes demonstrate the potential benefits of making targeted changes to model architecture, suggesting great promise for improving security features in digital communication tools.