1. Introduction to Computer Security ©2004 Matt Bishop Information Security Principles Assistant Professor Dr. Sana’a Wafa Al-Sayegh 1 st Semester 2008-2009 ITGD 2202 University of Palestine

2. Introduction to Computer Security ©2004 Matt Bishop Information Hiding with Steganography -Ashwini Tumne

3. Introduction to Computer Security ©2004 Matt Bishop Agenda What is information hiding? What is steganography? Steganography System Steganographic model Modern day applications Steganography in digital images Steganography in MP3 Conclusion

4. ©2004 Matt Bishop What is information hiding

5. Introduction to Computer Security ©2004 Matt Bishop What is Steganography? Embedding information in given media without making any visible changes to it

6. Introduction to Computer Security ©2004 Matt Bishop Steganography System

7. Introduction to Computer Security ©2004 Matt Bishop Steganographic model Stego Inverse Function f E -1 Stego Function f E cover C Embedded (E) Stego S Stego Key (K) Embedded (E) SenderRecipient The Stegosystem

8. Introduction to Computer Security ©2004 Matt Bishop Stego-system criteria Cover data should not be significantly modified ie perceptible to human perception system The embedded data should be directly encoded in the cover & not in wrapper or header Embedded data should be immune to modifications to cover Distortion cannot be eliminated so error-correcting codes need to be included whenever required

9. Introduction to Computer Security ©2004 Matt Bishop Steganography in images Way images are stored: Array of numbers representing RGB values for each pixel Common images are in 8-bit/pixel and 24-bit/pixel format. 24-bit images have lot of space for storage but are huge and invite compression 8-bits are good options. Proper selection of cover image is important. Best candidates: gray scale images.. Cashing on limitations of perception in human vision

10. Introduction to Computer Security ©2004 Matt Bishop Least Significant Bit method Consider a 24 bit picture Data to be inserted: character ‘A’: (10000011) Host pixels: 3 pixel will be used to store one character of 8-bits The pixels which would be selected for holding the data are chosen on the basis of the key which can be a random number. Ex: 00100111 11101001 11001000 00100111 11001000 11101001 11001000 00100111 11101001 Embedding ‘A’ 00100111 11101000 11001000 00100110 11001000 11101000 11001001 00100111 11101001 According to researchers on an average only 50% of the pixels actually change from 0-1 or 1-0.

11. Introduction to Computer Security ©2004 Matt Bishop Sample palettes Red color shade variations Drastic & Subtle shade variations Gray Scale shade variations

12. Introduction to Computer Security ©2004 Matt Bishop Masking and filtering Hide information by marking an image in a manner similar to paper watermarks. Watermarking techniques integrate a data in image Faint but perceptible signal is covered by another one that makes the first non-perceptible to human eye. No destruction of data with image compression. Used widely for digital watermarking and fingerprinting. Used for hiding a image within another.

13. Introduction to Computer Security ©2004 Matt Bishop Masking and filtering (contd.) Cover pixel : 01011100 Secret image pixel: 11101010 Resultant pixel: 01011101 (3+5) The contribution of each pixel can be varied to achieve desired effect.

14. Introduction to Computer Security ©2004 Matt Bishop + = Example: Copyright Fabian A.P. Petitcolas, Computer Laboratory, University of Cambridge http://www.cl.cam.ac.uk/~fapp2/steganography/image_downgrading/ 8-bit (256 grayscale) images. TOP SECRET

15. Introduction to Computer Security ©2004 Matt Bishop Sacrificing 2 bits of cover to carry 2 bits of secret image Original ImageExtracted Image

16. Introduction to Computer Security ©2004 Matt Bishop Sacrificing 5 bits of cover to carry 5 bits of secret image Original ImageExtracted Image

17. Introduction to Computer Security ©2004 Matt Bishop Algorithms and transformation Message is embedded into cover image by modulating the original coefficients in transform domain such as Discrete Cosine Transform or Wavelet transform. To encode: Take the DCT or wavelet transform of the cover image Find the coefficients below a certain threshold Replace these bits with bits to be hidden (can use LSB insertion) Take the inverse transform Store as regular image. To decode this image to get the message: Find the coefficients below a certain threshold Extract bits of data from these coefficients Combine bits into actual message

18. Introduction to Computer Security ©2004 Matt Bishop Steganography in MP3 Music company publishes albums in mp3 and publishes over internet. Some people take these mp3 files and publish under their own name. Case goes to court. The Music company needs to prove that the material which is exhibit is indeed the one they published. They need a hidden copyright.

19. Introduction to Computer Security ©2004 Matt Bishop Principle : Audio signals contain a significant portion of information that can be discarded without average listener noticing the change. MP3Stego – tool developed by Fabien A.P. Petitcolas Tool operates within MP3 encoding process The data to be hidden is first compressed, encrypted and hidden in MP3 bit stream. Quantization of original audio signal takes place. At the same time, for some selected points, data is introduced in the quantized output, Distortions introduced by these are constantly checked for to satisfy the psychoacoustics model. A variable records the number of bits that are for data in the actual audio, data for huffman coding and hidden data. Key is selected using pseudo random bit generator based on SHA-1 and dictates the values that would be modified to hold the hidden data. Steganography in MP3 (contd.)

20. Introduction to Computer Security ©2004 Matt Bishop Conclusion Steganography has its place in the security. On its own, it won’t serve much but when used as a layer of cryptography, it would lead to a greater security. Far fetched applications in privacy protection and intellectual property rights protection. Research is going on in both the directions –One is how to incorporate hidden or visible copyright information in various media, which would be published. –At the same time, in opposite direction, researcher are working on how to detect the trafficking of illicit material & covert messages published by certain outlawed groups.