A novel blind watermarking approach for medical image authentication using MinEigen value features

Abstract

Developing new watermarking approaches that consider special features of medical images become increasingly necessary. This paper proposes a new watermarking approach to ensure medical images authenticity, using MinEigen value features, chaotic sequence, and Quantization Index Modulation (QIM) in the spatial domain. The idea is to choose the 3 × 3 non overlapping blocks around MinEigen values points, then embed the watermark bits in these blocks using a novel blind way based on chaotic sequence and QIM. The proposed technique is purely blind and fast in terms of execution time. Experimental results demonstrate that the proposed approach is robust against all DICOM JPEG compression attacks while keeping high imperceptibility.