“Stego-Analysis Chain, Session Two” Novel Approach of Stego-Analysis System for Image File

A.W. Naji *, Shihab A. Hameed, Md Rafiqul Islam, B.B Zaidan, Teddy S. Gunawan, A.A Zaidan

Department of Electrical and Computer Engineering, Faculty of Engineering, International Islamic University Malaysia ,

Jalan Gombak, 53100 Kuala Lumpur, Malaysia. *e-mail: ahmed@iiu.edu.my

Abstract – Stenography systems has too many approaches , it has a lot of applications and different ways, too many type of attackers have been tried to attacked the information that already embed in the carrier, however there is not yet a real stego- analysis system offered, the goal of this chain is to have too many investigations on the stego analysis systems, this paper will focus on the image file, the earlier tests showing that we may use the histogram habits to classify the image in to image carried data hidden and original image, the study will focus in to two main idea, the first idea, on the colour image, the colour start become grouping parallel with the increasing of the data hidden and the second feature on the gray level image, the histogram in the original image was pure without colours , after the embedding method has been applied there are many colour start to appears as shown down in the histogram, we will try through this paper to offer a new stego-analysis system though study the habits of the histogram. (keyword): Stego-Analysis, Steganography, Hidden Data, Steganography Attackers, Conselmet Methodes

I. INTRODUCTION

Opportunities in the computer forensic career field are wide ranged and numerous, Computer forensic is not just law enforcement incorporated it is also used largely in the corporate world. With knowledge and training in the computer forensic area of study local, state, and federal law enforcement agencies, would be more likely considered you for career opportunities due to need of computer forensic examiners. More and more criminal acts are being investigated through computer, due to the increase of crimes being committed with the aid of computers. As long as computers are being widely used as today in society, there are always going to be a need for computer forensic examiners. The fields of finding evidence about data embedding in other media have been listed under the computer forensic; perhaps, some people use some technologies for awful things, finding evidence is an interesting filed, and many research nowadays running to allocate the bad people and stop them or at least know those plans. This study will focus on the data hidden on the image, in particular, the study will go through proposing a way to discriminate between the image that include the data hidden and the normal image[1][2][3].

A. Motivation

This work is extend to our work which have been done before, the work was improving a high rate data hidden and study the impact that growing on the image texture after increase the rate of data hidden, after this empirical study we note some of the new habits at the image texture, this habit may escort to implement a stego-analysis system, the work will be propose a new idea for implementing a new stego-analysis system through study the habits of the image texture before and after hidden data, we expect that we will define two new approaches for stego-analysis systems, one for gray level image and other for color image, perhaps the result will identify a very good result and maybe it will guide us to know the method of data hidden which apply on the image.[1][2], a study had offered from “NATIONAL SCIENCE AND TECHNOLOGY COUNCIL” at United State under the name “FEDERAL PLAN FOR CYBER SECURITY AND INFORMATION ASSURANCE RESEARCH AND DEVELOPMENT” in the middle of 2006 was also one of the motivation to approve such as this work, it’s clearly said as a conclusion of this study “There is no universally applicable methodology for detecting steganographic embeddings, and the few general principles that exist tend to be ad hoc. In cyberspace, steganography provides a capability for transmitting information undetected” [7].

B. Related work

A few research have done on this area, in this part we will review some work in the stego-analysis systems One research in this area under name “Is Image Steganography Natural?”, This paper experimentally investigates if stego-images, bearing a secret message, are statistically “natural.” For this purpose, we use recent results on the statistics of natural images and investigate the effect of some popular steganography techniques. We found that these fundamental statistics of natural images are, in fact, generally altered by the hidden “non natural” information. Frequently, the change is consistently biased in a given direction. However, for the class of natural images considered, the change generally falls within the intrinsic variability of the statistics, and, thus, does not allow for

2009 IACSIT Spring Conference

978-0-7695-3653-8/09 $25.00 © 2009 IEEE

DOI 10.1109/IACSIT-SC.2009.104

398

2009 International Association of Computer Science and Information Technology - Spring Conference

978-0-7695-3653-8/09 $25.00 © 2009 IEEE

DOI 10.1109/IACSIT-SC.2009.104

410

reliable detection, unless knowledge of the data hiding process is taken into account[5][6]. In the latter case, significant levels of detection are demonstrated. Other research is “Visual Steganalysis of LSB-encoded Natural Images”, Contemporary steganographic systems encode Hidden messages inside the least significant bit layers of color natural images. The presence of these messages is difficult to detect through statistical attacks. This study examined whether humans could detect steganography in natural images using a controlled 2AFC discrimination task. While d’>1 was observed for color layers 3-8, Layer 1 had a negative’. Thus, Layer 1 embedding is highly resistant to visual attack, since observers were more likely to indicate the presence of steganography in the control image than the embedded image [5][6].

II. METHODOLOGY

In the embedding methods there are many tricks and many techniques, in this research we proposed two approaches for stego-analysis systems depend on the histogram behavior, the 1st one for the color image and the 2nd one gray level image the system will classify image into image include data hidden and original image, Figure 1 showing the waterfall model for the methodology where the image in the 1st stage exposed to the embedding process, in the 2nd stage drawing the histogram for the algorithm, and then extract the feature of the histogram after the impact of the stego function on the image, the same process will apply on the gray level image as well as the color image [4][1].

A. Earlier Experimental Result

The earlier experimental result and the assessment test for this project had already approved, where the embedding operation have been in to two type of image (gray level image, color image), the new habits of the histogram shows in the color image, color become a group and distant peaks in other word the distant between summits In direct proportion with the increasing data hidden in the image, and this bizarre conduct is our start point to identify the new approach of the stego-analysis systems next figures showing the result of embedding operation also the histogram before hidden data and after hidden data

Figure 2. Image before Data Hidden

Figure 3. Image after Data Hidden The picture almost the same even there are more than 50% of its size as data hidden.

Figure 1. Waterfall Model

Import Image

Embedding Data

Make Histogram

Extract the feature

Analysis

Modulate the classification

399411

Figure 4. Showing the Summit and the Bizarre Conduct for the Histogram. As we know the histogram is the number of repetitions of each level of color, in other word how many time the level of the color x repetition in the image, where x between 0-256 if the image 24-bit.

Figure 4. Histogram before Data Hidden.

Figure 5. Histogram after Data Hidden

In the gray level image there is more than this feature, as we know the gray level image is the image that have the same value for the three color in each pixel, for example if pixel x has three color and the value of blue is 33 its mean each of the red and the green should be 33, so that we may get one of the gray level color, this feature guide us to define a way to implement a system detect the gray level image, where the embedding operation should give some result that affect the histogram, the new histogram of each image will give some level of color depend on the change of the three color , we will consider that if there is an embedding data there should be a change otherwise the data match with the picture and that is impassable Next figures showing the images before and after hidden data also the histogram before and after data hidden [4].

Figure 6. Image before.

Figure 7. Image after.

400412

Figure 7. Color in the Histogram of Gray Level Image.

Figure 8. Histogram before.

Figure 9. Histogram after.

III. CONCLUSION

Steganography is the art of hiding important information inside other innocuous media in an imperceptible way. this paper discuses the applicability of implement the stego-analysis system though extract the histogram features , the goal of this study depend on two feature the first idea, on the colour image, the colour start become grouping parallel with the increasing of the data hidden and the second feature on the gray level image, the histogram in the original image was pure without colours , after the embedding method has been applied there are many colour start to appears as shown down in the histogram, we will try through this paper to offer a new stego-analysis system though study the habits of the histogram, the result was good as shewn above.

ACKNOWLEDGEMENT

Thanks in advance for the entire worker in this project, and the people who support in any way, also I want to thank IIUM, UM for the support which came from them

REFERENCES

[1] B.B.Zaidan, A.A.Zaidan, A.Y.Taqa, Fazidah Othman “An Empirical Study for Impact of the Increment the Size of Hidden Data on the Image Texture”, ICFCC09, KL, Malaysia,2009.

[2] A.A.Zaidan, B.B.Zaidan, A.W.Naji, Fazida Othman & A.Taka “Securing Cover-File of Hidden Data Using Statistical Technique and AES Encryption Algorithm”, ICSAP09, KL, Malaysia,2009.

[3] B.B.Zaidan., A.A.Zaidan.& Fazida Othman, “Enhancement of the Amount of Hidden Data and the Quality of Image“Faculty MyEduSec08 University of Malaya, Kuala Lumpur, Malaysia,2008.

[4] B.B.Zaidan,A.A.zaidan, “Enhancement of the Size of Hidden Data and the Quality of Image Using LSB Algorithm”, Master research, Kuala Lumpur, Malaysia,2008.

[5] A.M, G.S, & G.L, “ Is Image Steganography Natural ” IEEE, USA, 2007.

[6] Paul A. Watters1, Frances Martin2, H. Steffen Stripf, “Visual Steganalysis of LSB-encoded Natural Images”, IEEE,USA, 2007.

[7] National Science and Technology Council, “Federal Plan for Cyber Security and Information Assurance Research and Development, USA, 2006.

[8] Dorothy, E.R, D.K, “Cryptography and Data Security”, IEEE International Symposium on Canada Electronics (ISKE), University of Canada, Canada, Vol.6, pp.119-122, 2006.

[9] Johnson, N. F. S. D, Z., “Information Hiding: Steganography and Watermarking-Attacks and Countermeasures”, Center for Secure Information Systems (CSIS), Boston/Dordrecht/London, George Mason University, 2006.

[10] Katzenbeisser, S. P., A. P, “Information Hiding Techniques for Steganography and Digital watermarking”, available from: Artech house pub, 2005.

[11] Katzenbeisser S. & Petitcolas, F. A., “Information Hiding Techniques for Steganography and Digital Watermarking”, Artech House, USA, 2001.

[12] Katzenbeisser, S. P., P. A,“ Information Hiding Techniques for Steganography and Digital water marking ”, Proceedings of the Eighth Symposium on programming Languages and Software Tools SPLST'05, USA, 2001.

401413