A Novel Secure Digital Watermarking Method Using All Phase Discrete Cosine Haar Biorthogonal Transform and Parrot Optimizer
Lokesh Kumar Verma, Ajeet Singh · International Journal of Image and Graphics · 2025
Securing or protecting digital content is crucial for business owners and entrepreneurs in today’s technologically advanced world. Through the use of copyright details, such as copyrights incorporated into the data, their digital content can be lawfully protected. This research presents a novel secure digital watermarking method that leverages the All Phase Discrete Cosine Haar Biorthogonal Transform (APDCHBT) for robust watermark embedding into host images. After splitting the host image into blocks; APDCHBT is used to break each block down into its component frequency parts. The watermark is embedded into the frequency coefficients, and the Parrot Optimizer (PO) optimizes scaling variables that determine the embedding strength. By maximizing the Normalized Correlation (NC) for robustness and the Peak Signal-to-Noise Ratio (PSNR) for imperceptibility, this optimizer iteratively finds the ideal scaling factors. The suggested approach exhibits strong resilience to assaults, upholding a harmonious equilibrium between imperceptibility and watermark security. Compared to conventional DWT- and SVD-based watermarking approaches, the proposed method achieves an average PSNR improvement of 12.3%, NC enhancement of 7.8%, and BER reduction by over two orders of magnitude, demonstrating superior resilience to common image processing attacks. The maximum obtained values for the metrics such as PSNR, SSIM, BER, and NC for the proposed research are up to 38 dB, 0.779, 10[Formula: see text], and 0.849.