Detecting the LSB-LFSR Algorithm Stego-Image Using the Statistical Analysis and Visual Attack

Abstract

The aim of this paper is to detect the hidden information in BMP imageswhen using Least Significant Bit (LSB) technique. In general, the passivesteganalysis system deals with some stego-tools based on three stages:Diagnosis, Breaking and Extraction. In this paper we focus in Diagnosisstage to attack a Linear Feedback Shift Register (LFSR) algorithm as stegotools. In the diagnosis process, the statistical analysis and visual test aresuggested as tools for detection. These tests are: Mean Square Error (MSE)for stego-cover images, and Laplace Operator, chi-square analysis andvisual test for stego only attack.