سامانه همایش های دانشگاه آزاد اسلامی واحد نجف آباد

صفحه اصلی / سومین کنفرانس بین المللی مهندسی و علوم کامپیوتر و اولین کنفرانس ملی سامانه های هوشمند، کلان داده و رایانش پیشرفته

A Comprehensive Review of LSB-Based Image Steganography in Grayscale Images

نویسندگان :

Tahereh Salkhordeh¹ Farzaneh Kaviani²

1- Isfahan (Khorasgan) Branch, Islamic Azad University, Isfahan, Iran 2- دانشگاه آزاد خوارسگان اصفهان

کلمات کلیدی :

Image Steganography،Least Significant Bit (LSB)،Grayscale Images،Imperceptibility،Steganalysis Resistance،Embedding Capacity،Information Security

چکیده :

Abstract—The rapid growth of digital communication over public networks has intensified the demand for secure and covert data transmission mechanisms. Image steganography addresses this challenge by concealing confidential information within visually innocuous digital images, thereby reducing the likelihood of detection. Among various embedding strategies, Least Significant Bit (LSB)–based techniques have been widely adopted for grayscale images due to their simplicity and high embedding efficiency. This paper presents a comprehensive review of LSB-based steganography methods for grayscale images. Thirteen representative techniques published between 2001 and 2025 are systematically analyzed and classified into basic LSB substitution, modified and improved LSB schemes, adaptive LSB-based methods, and pixel value differencing approaches. The reviewed methods are evaluated using commonly adopted performance metrics, including Peak Signal-to-Noise Ratio (PSNR), Mean Squared Error (MSE), embedding capacity, Structural Similarity Index (SSIM), and resistance to statistical and learning-based steganalysis. The analysis reveals a clear research trend toward enhancing imperceptibility and security rather than solely maximizing payload capacity. Although recent techniques demonstrate significant improvements in visual quality and robustness against conventional attacks, several challenges remain unresolved, particularly regarding robustness to compression, resistance to deep learning–based steganalysis, and energy-efficient real-time implementations. By identifying these open issues, this review provides valuable insights and guidance for the development of more resilient and practical LSB-based steganography systems for grayscale images.