A genetic algorithm based medical image watermarking for improving robustness and fidelity in wavelet domain

Abstract

In modern medical diagnosis, protecting the medical images from vulnerability attacks gains more importance. The proposed work focuses on providing better robustness, when the image has undergone various geometric attacks combined with balancing the trade-off between imperceptibility and robustness, by introducing genetic algorithm based watermarking method combined with Discrete Wavelet Transform (DWT) and Singular Value Decomposition (SVD). False positive errors due to singular values are balanced by calculating Key Component (KC). Authentication of watermarks for patient verification is done at the watermark extraction phase by decrypting the watermark through logistic map permutation method. The Peak Signal-to-Noise Ratio (PSNR) and Normalized Cross-Correlation (NCC) values calculate multi-objective fitness values of the chromosome. The proposed method shows the better robustness and security to the medical images along with authentication to the watermark.