A High-Capacity Steganographic Method Using Hamming Distance

Abstract

The process of hiding information in digital media so that an observer cannot possibly find it is known as steganography. With the growing need for secure communication channels in the digital age the topic has taken on greater significance. By hiding the hidden information behind innocuous cover images steganography ensures that only those who are aware of the extraction technique can access it. Among many techniques the Least Significant Bit (LSB) method is a simple and widely used approach. However, this technique usually causes noticeable distortions in the cover image which raises the risk of being discovered. Our proposed solution uses an adaptive data embedding strategy based on Hamming distance calculations to overcome the distortion problem. Less significant bits (LSBs) of the cover image pixels and the bits containing the secret data are two binary strings. The difference between them is measured by the Hamming distance. Finding the Hamming distance between the four least significant bits of each pixel in the cover image and the four bits of confidential data is the first step in the embedding process. All four of the confidential information bits are inserted when the distance is short, indicating a minimum distinction. Alternatively, fewer bits are inserted to reduce the impact on the original image when the Hamming distance is significant. To be more exact when the distance is big, we look at the 3-bit distance first then the 2-bit distance and so on. The overall distortion is reduced because the hierarchical technique ensures that the embedding process adapts to the unique features of the cover image. We conducted experiments on frequently used public datasets to confirm the efficacy of our method. The visual fidelity of the steganographic images and the total amount of data that could be included were the two main metrics we used to evaluate the performance. While none of the other methods achieves more than 1 bpp, the proposed method outperforms other state-of-the-art methods in achieving a maximum of 2 bpp while maintaining acceptable marked image quality with PSNR values well above the nominal threshold of 30 dB. These improvements show how our approach can strengthen steganographic systems robustness and security. In brief our approach to adaptive data embedding which makes use of Hamming distance computations offers a viable solution to the distortion problem in LSB steganography. We have achieved a balance between the amount of information that can be incorporated and the image quality by adjusting the number of bits administered dynamically. This development advances secure data hiding strategies.