A Reversible Medical Image Watermarking Scheme for Advanced Image Tampering Detection

Medical image protection against tampering attacks remains an ongoing issue, despite the numerous watermarking-based solutions developed thus far. This is partly due to the fact that existing schemes often overlook the potential application of sophisticated tampering attacks specifically designed to exploit the strengths and weaknesses of the protective mechanisms. In this paper, we present an attack targeting least significant bit (LSB) or quantization-based fragile watermarking systems. This tampering attack alters the image while preserving an inconspicuous image component containing a fragile watermark. To counteract this attack, we propose a quantization-based fragile watermarking method that exhibits resistance to advanced tampering attacks by incorporating variations in quantization steps. The robustness of the proposed method is supported theoretically and validated through experimental analysis. For practical implementation, the proposed fragile watermarking method is employed within the region of interest (ROI). Distortions introduced during the embedding of the fragile watermark are further incorporated into the region of non-interest (RONI) to ensure the reversibility of the scheme. The developed method was thoroughly investigated across a wide range of parameter values and external factors, demonstrating exceptional localization accuracy for both conventional and advanced tampering attacks. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd 2024.