• Hari Balaji J C
• Saikrishna H
• Anitha M

This paper presents SecurePix, a robust stegware detection framework that integrates a Variational Autoencoder (VAE) and a Conditional Generative Adversarial Network (cGAN) for image steganalysis. The VAE performs unsupervised anomaly detection by learning the statistical distribution of clean images and identifying stego images through reconstruction errors. The cGAN discriminator is repurposed as a supervised classifier to accurately distinguish clean and stegware images. Experimental analysis demonstrates that the hybrid approach significantly outperforms single-model baselines, achieving high accuracy and reliability in detecting both known and subtle steganographic threats.

Stegware Detection, Image Steganalysis, Variational Autoencoder (VAE), Conditional GAN (cGAN), Anomaly Detection, Deep Learning, Malware Detection, Generative Models, Cybersecurity

"SecurePix: Stegware Detection using VAE – cGAN", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.13, Issue 1, page no.b205-b208, January-2026, Available :http://www.jetir.org/papers/JETIR2601133.pdf

"SecurePix: Stegware Detection using VAE – cGAN", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.13, Issue 1, page no. ppb205-b208, January-2026, Available at : http://www.jetir.org/papers/JETIR2601133.pdf