Enhancing Image Security through PCA-IWT-Based Nil Steganography \\under Distortion Scenarios

Abstract

Security has become a paramount concern across various fields of study today, with particular emphasis on steganography mechanisms as sought-after and promising research areas aimed at safeguarding multimedia data from unauthorized access. This study introduces a robust nil steganography technique that enhances its undetectable nature and resilience to various image processing attacks by combining Principal Component Analysis (PCA) with the Integer Wavelet Transform (IWT). The host image is converted to grayscale, the LL band is extracted using IWT, and the image is divided into non-overlapping $4 \times 4$ blocks as part of the concealment process. Principal features are then extracted from each block using PCA, and these are then XORed with the binary emblem to produce the confidential share. To produce the confidential share, three emblems are tested, and the XOR operation is performed to combine each emblem with the extracted features. According to experimental results, the system performs well in terms of Peak Signal-to-Noise Ratio (PSNR), Normalized Correlation (NC), and Bit Error Rate (BER) across various attack scenarios, including histogram equalization, Gaussian noise, and JPEG compression. In contrast to current techniques, the suggested method exhibits better imperceptibility and robustness. The algorithm can be considered suitable for real-world applications in secure image processing because it maintains a computational complexity of $\mathrm{O}\left(\mathrm{N}^{2}\right)$.