Abstract—This paper presents a hybrid blind watermarking scheme that integrates 2-level Discrete Wavelet Transform (DWT), Discrete Cosine Transform (DCT), and Singular Value Decomposition (SVD), with embedding strength optimization via the Cheetah Optimization Algorithm (COA). The proposed method achieves exceptional imperceptibility with Peak Signal to Noise Ratio (PSNR) values consistently above 49 dB (reaching up to 57 dB), representing 8–30% improvements over existing methods, while maintaining Structural Similarity Index Measure (SSIM) values above 0.995 and Normalized Cross Correlation (NCC) values near unity (≈1.0000). Comprehensive robustness evaluation demonstrates superior performance against diverse attacks, particularly excelling under geometric distortions and enhancement-based attacks such as histogram equalization (NCC: 0.9902 vs. 0.72 for baseline methods) and sharpening (NCC: 1.0000). The process maintains a moderate watermark capacity of 0.50–1.00 bits per pixel, deliberately optimized for hardware feasibility rather than maximum payload. With reduced computational requirements due to smaller watermark dimensions, the approach achieves embedding and extraction times of 0.2321 s and 0.0227 s, respectively, making it suitable for resource-constrained environments. The COA optimization automatically determines the optimal embedding strength, balancing imperceptibility and robustness. The proposed framework demonstrates consistent superiority across multiple benchmark comparisons, making it highly suitable for real-time applications and future Field Programmable Gate Array (FPGA)/Application Specific Integrated Circuit (ASIC) hardware implementations in secure medical imaging and multimedia protection systems.
 
Keywords—Discrete Wavelet Transform (DWT), Discrete Cosine Transform (DCT), Singular Value Decomposition (SVD), Cheetah Optimization Algorithm (COA), Peak Signal to Noise Ratio (PSNR), Normalized Cross Correlation (NCC)


Copyright © 2026 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).