Towards a High-capacity Data Concealment for Spatial Domain Image-steganography
Ntivuguruzwa Jean De La Croix, Moh Rosy Haqqy Aminy, Diajeng Alfisyahrinnisa Anandha, Hammuda Arsyad, Muhammad Nevin, Tohari Ahmad · 2023
In the digital age of internet-based communication, safeguarding data confidentiality faces heightened challenges from unauthorized access and network policy breaches. To alleviate these threats, there is a growing focus on employing data-hiding techniques, such as the steganography approach emerging as the most effective among various data protection algorithms. Traditionally, steganography used the concept of difference expansion (DE) for reversible integer transformation to insert the data into images. However, in specific scenarios, the existing DE falls short in enhancing the visual fidelity of resulting stego images. This paper introduces a novel DE-based scheme that conceals secret data in transformed differences between neighboring pixels, addressing this limitation. Experimental results reveal a substantial 21-decibel improvement in Peak Signal-to-Noise Ratio (PSNR) and an increased embedding capacity with around 49,039 more embeddable pixels than existing methods. Notably, achieving a PSNR of 69.73 dB surpasses the 30 dB threshold for visible stego images.