A Secure Hybrid Watermarking Framework using DWT, SVD and Elliptic Curve Cryptography
Kosuru Asha, S.V.K.S. Pratyusha, P. Satyanarayana Murty · 2025
We introduce a novel and secure hybrid digital image watermarking framework in this project. This framework incorporates Discrete Wavelet Transform (DWT), Singular Value Decomposition (SVD), and Elliptic Curve Cryptography (ECC) in order to improve the robustness and security of the watermarking process. This method involves dividing the cover image into $\mathbf{N} \times \mathbf{N}$ blocks, and then generating a reference image with a size of $\mathbf{p} \times \mathbf{p}$ blocks. This is accomplished by computing the number of edges included within each block, and then arranging the blocks in ascending order to get the median edge. This median is utilized in the process of generating the reference image, which is initially processed by means of DWT, and then SVD is applied. A cryptographic method is utilized to encrypt the watermark image, which is then subjected to SVD processing before being encoded and included into the reference image. The hybrid technique has been shown to provide good imperceptibility, with low distortion to the host image, while preserving strong resistance against standard image processing attacks such as noise, cropping, and compression, as demonstrated by the results of the experiments. The framework is especially well-suited for applications that require both robustness and security, such as those that require protection of copyright and authentication of digital media. This hybrid watermarking technology beats existing methods in terms of Peak Signal-to-Noise Ratio (PSNR), Normalized Cross Correlation (NCC), and robustness against a wide range of attacks, which makes it a very effective option for secure medical image watermarking. Our findings demonstrate that this hybrid watermarking technique is superior to previously used methods.