A robust blind hybrid watermarking technique for medical images based on contourlet and dual-tree complex wavelet transforms
Year: 2025
Type: article
Abstract: This paper proposes a blind hybrid watermarking scheme that addresses the critical need for data security and integrity in e-health systems. The method integrates the Contourlet Transform (CT) and Dual-Tree Complex Wavelet Transform (DTCWT) to achieve a robust and imperceptible watermarking solution. The host image is decomposed using CT, after which the low-frequency sub-band is further processed with DTCWT. A differential embedding technique then embeds the watermark into the high-frequency coefficients using a secret key, enabling blind watermark extraction without the original image. Experiments on X-ray, MRI, CT, and ultrasound images demonstrate consistent performance across modalities. The method maintains high imperceptibility PSNR > 43.58 dB, SSIM > 0.97) and strong robustness against 26 different attacks. Compared with existing approaches, it improves imperceptibility by up to 11% and robustness under common attacks by 10–50% (normalized correlation gain). With blind extraction capability, low computational cost, and proven robustness, the proposed scheme is well-suited for secure medical image transmission in real-world telemedicine applications.
Authors: Ali Kouadri, Ali Benziane, Abdelhalim Rabehi, Abdelaziz Rabehi, Takele Ferede Agajie, Abdullah K. Alanazi
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $1,790
OpenAlex ID: W4417294879