An EMD-based data hiding scheme for JPEG images. Issue 3 (3rd July 2021)

• Record Type: Journal Article
• Authors: Xie, Xiao-Zhu; Chang, Chin-Chen; Horng, Ji-Hwei
• Is Part Of: Connection science. Volume 33:Issue 3(2021)
• Page Start: 515
• Page End: 531
• Publication Date: 2021-07-03
• Subjects: High embedding capacity -- JPEG image -- data hiding -- EMD -- frequency domain; Neural computers -- Periodicals; Artificial intelligence -- Periodicals; Cognitive science -- Periodicals; Connectionism -- Periodicals; 006.3
• DOI: 10.1080/09540091.2020.1853055
• Languages: English
• ISSNs: 0954-0091

Abstract

Abstract : As a commonly used image format, JPEG image is popularly used as the carrier for data hiding. After analyzing the virtue of the exploiting-modification-direction (EMD) scheme, which is designed for the spatial domain, this paper reforms its extraction function and applies it to the non-zero alternating current (AC) coefficients of JPEG images. Besides, a block and coefficient selection strategy is utilised to further improve the image quality and suppress the increase of file size. More specifically, the smooth discrete cosine transform (DCT) blocks with fewer zero AC coefficients are preferentially selected to embed data. The experiments indicate that our scheme achieves a high embedding capacity, meanwhile, the stego-image quality keeps good and the file size preserves well, moreover, our scheme is superior to the state-of-the-art scheme.