Bit-4 of Frequency Domain-DCT Steganography

Technique

1

Nedal M. S. Kafri and Hani Y. Suleiman

Networked Digital Technologies, 2009. NDT '09. First International Conference on

Department of Computer Science & IT,Al Quds University, Palestine.

Outline

• Introduction

• Background

• Proposed algorithm

• Experimental results

• Conclusion

• Reference

2

Introduction

• “Steganography si the art of hiding information in a way that prevent the detection of hidden message” defined by Neil Johnson

• The most common image format include BMP, GIF and JPEG.

• Generally, the main idea of the proposed method is to hide message bits in a significant bit of a transform domain coefficients of an image and inverse if to spatial domain.

• The most well-known transform coding techniques used to implement lossy image compression are the discrete cosine transform(DCT) and Wavelate.

3

Background

• Most of the camouflage process use the redundant bits of an image to embed secret message.

• In RBG images, the blue one is the most impercetptible to human eye.• Image steganography shcemes can be divided into two domain:

– Spatial & Frequency• The proposed method is hiding in redundant bits on frequency domain.

4

Background(cont.) SSB-4(System Steganography Bit-4) In RBG images, the blue one is the most impercetptible to human eye. Image steganography shcemes can be divided into two domain:

– Spatial & Frequency The proposed method is hiding in redundant bits on frequency domain. Imbedding data in the transform domain, the hidden data resides in more

robust areas, spread across the entire image, and provides better resistance against signal processing

5

Background(cont.) The most common frequency domain methods usually used in image

processing are the 2D DCT and Wavelet PSNR (Peak Signal to Noise Ratio) is a technical approach usually used to

evaluate the real quality of stego image Typical values for the PSNR in image and video compression are between

30 and 50 dB, where higher is better.

6

Proposed algorithm

Step1: Partitioned the image into 8X8 blocks(F(x,y)), then calculate the 64 DCT Coefficients S(u,v)

Step2: Embedding message bits– if(S(u,v) ≠ 0 && c ≠ message bit){

4th bit = message bit and modify the bits 1st, 2nd, 3rd and/or 5th}

Step3: Apply the IDCT converting S'(u,v) to F'(x,y)

7

Experimental Results

• Comparing three methods(LSB, LSB&DCT, Bit4&DCT) by using PSNR

8

Conclusion

• The obtained experimental results indicate the proposed method will be a good and acceptable steganography scheme.

• By imbed-ding information in the main significant bits of the DCT domain, the hidden message resides in more robust areas, spread across the entire stego image, and provides better re-sistance against stiganalysis process than other techniques.

• Though the paper didn't show the capacity of the proposed method(due to the fact that it just hidden 1 bit in the image), we could get the idea that if using LSB plus DCT method the image PSNR will raise greatly.

• In this paper we could easily understand the 4th bit DCT image steganography method due to the clear discreption. If we're going to write about DCT steganography method we could refer this paper.

9

Reference N. F. Johnson, S. Jajodia,“Exploring Steganography: Seeing the Unseen,”IEEE

Computer, 31(2): 26-34 (1998) T. Morkel, J. Eloff, and M. Olivier, “An overview of image steganography”, In

Proceedings of the Fifth Annual Informa-tion Security South Africa Conference (ISSA2005), (Sandton, South Africa, Jun/Jul. 2005).

J. Rodrigues, J. Rios, and W. Puech “SSB-4 System of Steganography using bit 4”, In International Workshop on Image Analysis for Multimedia WIAMIS, (Montreux, May 2005).

10