Secure Audio Embedding in Images with Image Transforms and Decompositions
Shyam Tripathi, Kritika Agrahari, Ashwini K M · 2025
Information technology sees unparalleled growth. The demand for stronger data security relating to some sensitive class of multimedia data is growing. This article compares creative methods of embedding audio into images using various image transforms and decomposition methods namely DWT, LP, and PCA. This approach ensures augmented security, robustness, and imperceptibility, and addresses known pitfalls in standard interleaving techniques. With DWT performing hierarchical multi-resolution analysis, LP capturing multi-scale representation, and PCA enabling dimensionality reduction, the embedding process is optimized to preserve both audio and image quality. Rigorous experimentation under attack scenarios ensures successful embedding and secure transmission. Objective evaluations like SNR, PESQ, BER, and LSD affirm robustness and perceptual transparency. The work demonstrates notable improvement, making it a viable solution for secure multimedia communication.