Cyber Sentinel: A Combined Steganographic and Encryption Framework for Enhanced Data Privacy

Abstract

In contemporary times, the transmission of information through insecure channels poses a significant risk due to the omnipresence of potential online intruders. To mitigate this concern, the conventional solution involves the implementation of encryption for safeguarding data. This study proposes an innovative strategy by integrating both steganography and cryptography to achieve dual-layered data encryption. The presented model leverages sophisticated artificial intelligence to categorize images into sensible, nonsensical, and suspicious classifications. By employing image saturation and segmentation techniques, concealed data can be unveiled. The model modifies the least significant bit of each pixel, incorporating covert data to generate a securely encrypted image. Adhering to the least significant bit (LSB) model and employing symmetric-key encryption, the resultant image maintains an indistinguishable appearance to an unencrypted steganographic photograph, both to the human observer and a computer lacking a stego-image for comparison. Importantly, the encrypted data remains impervious to decryption, even when subjected to image saturation or data extraction methods. This methodology augments cybersecurity, particularly in the realm of the Internet of Things (IoT), by furnishing a secure and inconspicuous means of data transmission.