A high payload block-based data hiding scheme using multi-encoding methods

With the rapid development of the Internet, large amounts of data are being transferred globally. As data are becoming increasingly valuable, securing the communicated information is a major concern. In data hiding, the secret data are concealed in a cover medium, which does not reveal the presence of the secret data. The least-significant-bit (LSB) substitution method is a popular data hiding scheme due to its high payload embedding capacity and low distortion. Several studies have achieved a high payload embedding capacity by combining the LSB substitution method with other techniques. A few studies employed an edge detection technique based on the sensitivity of the human visual system to embed the secret bits stored in the edge pixels. However, problems associated with the storage of edge information tend to reduce the embedding capacity. To increase the embedding capacity, a most-significant-bit (MSB) image with edge detection was employed in another study, wherein a fuzzy technique was applied to the edge matrix to increase the weak edge pixels. However, this approach did not ensure the generation of true edge pixels. In the meantime, a high payload reduces the quality of the stego-image. This study proposes a block-based data hiding scheme using a multi-encoding method. Compared to Bai et al.'s approach utilizing image features and fuzzy theory, the proposed method further improves performance by using block segmentation and optimization with multiple encodings. The experimental results show that this approach can enhance security at low cost and yield superior performance compared to Bai et al.'s method. After applying the proposed method, the payload has increased by an average of 396 bits, and the PSNR has increased by an average of 1.5286 dB compared to Bai et al.'s method.