Detection of LSB steganography via sample pair analysis

  • Authors:
  • S. Dumitrescu;Xiaolin Wu;Zhe Wang

  • Affiliations:
  • Dept. of Electr. & Comput. Eng., McMaster Univ., Hamilton, Ont., Canada;-;-

  • Venue:
  • IEEE Transactions on Signal Processing
  • Year:
  • 2003

Quantified Score

Hi-index 35.68

Visualization

Abstract

This paper introduces a new, principled approach to detecting least significant bit (LSB) steganography in digital signals such as images and audio. It is shown that the length of hidden messages embedded in the least significant bits of signal samples can be estimated with relatively high precision. The new steganalytic approach is based on some statistical measures of sample pairs that are highly sensitive to LSB embedding operations. The resulting detection algorithm is simple and fast. To evaluate the robustness of the proposed steganalytic approach, bounds on estimation errors are developed. Furthermore, the vulnerability of the new approach to possible attacks is also assessed, and counter measures are suggested.