Medical Image Watermarking Employing Discrete Wavelet Transform—HD–SVD Techniques

Digital watermarking is essential for research and diagnostic applications in the field of medical picture security and integrity verification. To increase resistance against various attacks, this study introduces a novel watermarking technique that combines Heisenberg decomposition with DWT-SVD. Using the LL sub-band, which is obtained by R-level DWT decomposition, the approach strategically embeds watermarks to take advantage of the sub-band's resistance to filtering and compression assaults. Hessenberg decomposition of the LL sub-band and SVD on the resulting upper Hessenberg matrix constitute the watermark embedding process. Watermark embedding is facilitated by adjustments to singular values, and the watermarked image is obtained by using the inverse DWT. Hessenberg decomposition, SVD, and decomposition of the watermarked image are all applied during the extraction process to recover the watermark. The method ensures imperceptibility while demonstrating resilience to frequent attacks. Metrics like as PSNR, SSIM, and NC are used to evaluate performance on medical pictures, and the results are encouraging. The technique produces high PSNR and SSIM values for a range of image sizes, suggesting significant imperceptibility. Moreover, the extracted watermark's durability is highlighted by the normalized correlation values. By presenting a novel approach to medical image watermarking, this study advances the development of reliable and safe diagnostic imaging systems.