A new approach to video steganography using pixel pattern matching and key segmentation

Abstract

Hiding data or information by embedding in a redundant medium like text, audio or video is called as steganography. In ancient time, different techniques such invisible ink, wax tablets, Morse code, etc. were used to send secret information from one place to another. Now in the epoch of digitalization, even the medium for steganography has evolved from paper, wood, etc. to multimedia. Video Steganography is a secure method to embed information in the video frame, but the main drawback of this system is that storing a significant amount of data introduces distortion in the pixels of the frame gets altered which increases the suspicion about the secret data present in it. Traditional techniques used Advance Encryption Standard to encrypt and Least Significant Bit to embed information in the image by modifying the least significant bit of the pixel. However, this introduces distortion in the image which can be recognised by the intruder. Least Significant Bit is not preferred if a substantial quantity of data is to be embedded. In the proposed system, the data is encrypted using Advance Encryption Standard and divided using arithmetic division method. In this approach, the data will be stored in the form of divisor, quotient & remainder. The location key is also distributed, encrypted and stored in different frames. Along with this pixel, pattern matching is also used to avoid distortion of the video frame. This system will be difficult to crack since the location key is divided as well as encrypted and stored in different video frames along with this the secret message is stored in the form of a quotient, a divisor and a remainder. Even if the system is attacked the chance of the intruder to predict the pattern will be difficult as the secret data is embedded with dual protection.