Curvelet Denoising Firewall for Universal Removal of Steganographic Image Content

• A Stego Firewall framework is proposed for universal removal of hidden steganographic content in digital images. • The method integrates Gaussian blur degradation, curvelet-based denoising, and image enhancement for effective stego destruction. • Achieves 85–90% elimination of embedded content while preserving high visual quality, validated by PSNR and SSIM metrics. • Demonstrates blind and algorithm-independent performance, requiring no prior knowledge of steganographic techniques. • Offers potential deployment as a cybersecurity defense module within firewalls and antivirus systems for secure image transmission. Steganalysis methods predominantly focus on detecting hidden content in images, while the embedded payload often remains intact, posing security risks. To address this limitation, this work proposes a blind and quality-preserving framework for steganographic content destruction that suppresses hidden payloads without compromising perceptual image quality. The method follows a three-stage pipeline consisting of: (i) controlled Gaussian blur–based degradation to weaken high-frequency embedding artifacts, (ii) curvelet-based denoising to suppress residual steganographic perturbations while preserving structural features, and (iii) image enhancement to restore visual fidelity. The approach is evaluated on a dataset of 512 × 512 grayscale images, including clean and stego images generated using seven representative spatial- and frequency-domain embedding algorithms at multiple payload levels. Performance is benchmarked against wavelet-based denoising and validated using nine steganalysis tools. Experimental results show an average payload suppression of 85–90%, while maintaining PSNR values above 35 dB and SSIM values close to 0.99, indicating minimal perceptual degradation. Although performance may vary at extreme payload densities, the results demonstrate that the proposed method is an effective algorithm-independent stego sanitization strategy with potential applicability in network-level image security systems.