High capacity data hiding for binary image based on block classification

Abstract

Data hiding is a technique by which secret data can be delivered securely by embedding it into a cover multimedia document. In this paper, a high capacity data hiding scheme based on block classification is proposed for binary images. In the proposed scheme, the block classification process determines complex regions in the image used for embedding secret data. For each block in the complex region of the image, some secret bits are embedded by lightly modifying few pixels to minimize an embedding distortion. On the receiver side, the secret bits can be then extracted without requiring the original image. Experimental results demonstrated that the proposed scheme obtained a high embedding capacity while maintaining a low distortion. In addition, when compared with previous schemes for binary images, the proposed scheme improved the embedding capacity significantly with the same level of embedding distortion.

Reversible data hiding in encrypted binary image with shared pixel prediction and halving compression

“…Although the embedding capacity is limited, Wang's scheme can still achieve better performance with the same embedding distortion. Furthermore, Nguyen et al [22] proposed a high embedding capacity data hiding scheme in binary image based on block classification. With a block classification process, this scheme identifies the complex image blocks, in which only a small number of pixels are embedded secret bits to minimize embedding distortion.…”

“…In this subsection, we carry out a series of experiments to compare the proposed scheme with five existing RDH schemes in binary image, Zhang's method [20], Ren's method [24], Nguyen's method [22], Wang's method [23], and Yin's method [21]. Notably, in these RDH Table 3 Scrambling analysis with GDD measure for the proposed scheme and Ren's scheme [24].…”

“…Fig. 12 Overall performance comparison between PSNR and embedding rate for six different RDH schemes in binary image Table 4 Average maximum embedding rate (MER) for six RDH schemes by using the large-scale image database BOSSBase Scheme Proposed Ren [24] Zhang [20] Nguyen [22] W a n g [ 23] Y i n [ 21] MER (bpp) 0.806 0.736 0.006 0.305 0.151 0.003…”

“…Although a considerable amount of RDH literature has been published on binary images, such as [20][21][22][23]. However, in these literatures, the developments mainly focus on plaintext binary images, while for encrypted binary images, too little RDH work has been devoted [24].…”

A Pattern-Based Password Authentication Scheme for Minimizing Shoulder Surfing Attack

“…The user needs to select the grids as their chosen password. The user can choose whatever pattern or styles they want as they wish [30]. There is no limit for the user to select how many grids they like to choose their password.…”