Image Sterilization to Prevent LSB-based Steganographic Transmission

TL;DR

This paper introduces an image sterilization technique to remove steganographic data embedded via LSB methods, achieving a 76% to 91% removal rate of stego pixels, enhancing data security.

Contribution

It is the first to propose and demonstrate a method for sterilizing images to eliminate LSB steganography, inspired by biological sterilization concepts.

Findings

Successfully sterilized 76% to 91% of stego pixels

Effective in removing embedded LSB steganographic data

Potential to improve image data security

Abstract

Sterilization is a very popular word used in biomedical testing (like removal of all microorganisms on surface of an article or in fluid using appropriate chemical products). Motivated by this biological analogy, we, for the first time, introduce the concept of sterilization of an image, i.e., removing any steganographic information embedded in the image. Experimental results show that our technique succeeded in sterilizing around 76% to 91% of stego pixels in an image on average, where data is embedded using LSB-based steganography.