COMBINATION OF JPEG ALGORITHM AND WATERMARKING FOR INCREASING THE QUALITY AND COMPRESSION RATIO - paper presenation

COMBINATION OF JPEG ALGORITHM AND WATERMARKING FOR INCREASING THE                          QUALITY AND COMPRESSION RATIO

1Muhammad Subali
2 Sarifuddin Madenda

              1 Gunadarma University(subali@staff.gunadarma.ac.id) 

2 Departement d’informatique et d’igenierie,UQO,Quebec(m.sarifuddin@uqo.ca)

Abstract

Jpeg formation is compression model of color image which is establish as an ISO and usually public

used. Jpeg compression technique started by RGB to YCbCr colour transformation process. And each color component Y, Cb, and Cr compressed by Discrete Cosine Transform (DCT) process,. Quantizing and Coding step by step. Bits of result coding from each component of Y, Cb and Cr combined one group in series.This paper explained compression technique can be optimized JPEG result compression by bit insertion of coding result from one color component to matrix quantization from another color component. The experiment shows that the same compression quality compression technique which develop in this paper have ratio of compression about 10.99% until 154% better then JPEG codec, and for the same compression ratio this technique give better quality compression. Keywords:Image Compression, JPEG,Watermarkning,MIG

1 Introduction

Technology develop of image acquatition more sophisticated, like digital camera, scanner, has produce color image until 8 Mpixel. The higher pixel will increase need of memory and time image transmission. Algorithm develop and optimize image compression technique is prominent solution for increasing transmission efficiency and data memory without decrease of image quality. Compression algorithms developed for loss less and lossy compression.

JPEG 2000 which use in DWT (Digital Wavelet Transform  is better product compression ratio then quality of JPEG. But its not used in the public. In another hand, JPEG used by DCT is loss less compression usually used by public and used by all the device of image acquisition. JPEG format acknowledge as ISO (International Standard Organization) for digital color image compression in 1992 (ISO/IEC JTC1 Dan CCITT Rec 81)[4].Base on digital image needs and copyright protection has develop watermarking introduce by G.C Langelar in 1996[16]. This method is insert the information in to picture by LSB of pixel.

This paper explain about combination of JPEG compression technique and watermarking to optimize compression ratio.

2 JPEG

It has been known, digital color image is pixel

collection where each pixel is color vector element three dimension (3-D) there are RGB, YCbCr 

etc[32]. Image with color representation YCbCr or 

YUV is used in JPEG compression. Figure 1 shows 

JPEG compression process, and figure 2 show general figure of bit collection of coding result in three

color component there are Y, Cb and Cr.

For each color component , JPEG divide image in to blocks 8 x 8 pixels. Then each pixel block will causes successively DCT process, Quantization, and Coding for color component Y, Cb and Cr. Bits collection of coding result from three color component is compressed image data. And bit header collection,tables and compressed image data formed JPEG bit stream.

To the JPEG image compression, conducted return process wich start by bits divide in each color

component and continue successfully by decoding process, de-quantization and DCT invers.This process shows in figure 3. In compression process and decompression process,quatization table and coding (entropy,Huffman coding) was a same.There for this table must participated to JPEG image

compressed[35].

3 Digital Watermarking

Digital watermarking is information insert process to audio signal, image or digital video. Information

insert has visible and invisible characteristic. In watermarking visible,information can be seen visual.

Specially for information like text or owner identi-fication logo from image or video. In watermarking invisible,information inserted to audio ,imageor video but cannot be seen by visual.Important application from this type is to copyright protection system. Steganography is one of part of invisible digital watermarking, where two person can communicated secreat by information insertion into audio, image and video.

One of simple methode invisible digital watermarking is bit insertion in to pixel by Least Significant Bit (LSB).To hide the information in to image LSB pixel, can be save 3 bits in each pixel.Manipulation of LSB pixel is quick and easy manner to hide the information. Invisible methode of digital watermarking from LSB use from modifi- cation technique of JPEG compression.

4 Develop Method

To increase compression ratio,in this part we suggested technique modification of JPEG compression

by collection between JPEG compression technique and watermarking. Two collection show in figure

4. Its different with JPEG compression technique in figure 2. This technique, bit insertion of coding

result color Cb component into quantized matrix of color Cr component. Then quantizing, inserting

and decoding in matrix Cr. Result bits of coding component Cr inserted again into quantized matrix

component Y. Finally, quantized and inserted matrix return coding and producing data bits of image

compression image. Data bits collected by header bit and quantizing and coding tables become bitstream which saved in file with extensi MIG/.MIG (Multimedia Informamation and data compression

Group).

Bit insertion conducted by using watermarking method like explained by G.C Langelar[16], is insertion of LSB in each matrix component.

5 Result Analysis

To proved performance this methode which developing in this paper,we do experiments with use comparing software are Adobe Photoshop CS2 versi 9.0.1, Morgan JPEG toolbox V2, Morgan JPEG 2000 Toolbox, ACD See 9 toward three standars images in bmp format (Babbon.bmp, Lenna.bmp, danpeppes.bmp) by resolution 512x512 pixel ( file size from each image is 769 kb).

In analyse experiment result conducted in two perspective. First, equal value ratio compression are grouped. So, PSNR value are analyzed which represented as comparation curves between compression ratio ( X axis ) and PSNR (Y Axis). The results are represented in the following figure 5 , 6 and 7. Second, equal PSNR values are grouped and then compression ratio analyzed, represented , as a comparison curves between PNSR ( X axis ) and comparison ratio ( Y axis). The results are represented in the following figure 8 , 9 and 10

Based on the three figure 5,6 and 7 shows that from equal compression ratio, MIG compression program has a high PSNR value toward adobe,Morgan, and ACD see as comparing software.

For peppers.bmp image with compression ratio about 2.7 until 16.7, MIG have PSNR value better then ACD See (smallest value). And for peppers.bmp image with compression ratio about 16 until 32, MIG have PSNR value 85.415% better then Morgan V2 (higher value). For baboon.bmp image with compression ratio 10.3 until 16, MIG have PSNR the worst value about 2 % from ACD

See. For Lena.bmp image with compression ratio about 22 until 35, MIG have PSNR the worst value

about 8.23% from ACD See.

Figure 8,9 and 10 show analysis that from equal PSNR value, All MIG compression ratio be better then comparing software. For papers image in PNSR value in above 41 dB. MIG have compression ratio about 10.99 % higher than ACD see (smaller increase in compression ratio). The higher compression ratio about 154% in MIG versus Morgan V2, And for baboon image about 36.2 dB of PSNR value. And for papers image of PNSR value in under 40 dB. MIG compression ratio about 3.8 % is smaller then ACD see.

6 Conclusion

MIG compression methode by collection of JPEG compression technique and Watermarking developed in this paper gives better result then general software, there are Adobe Photoshop CS2 versi 9.0.1, Morgan JPEG toolbox V2, Morgan JPEG 2000 Toolbox, ACD See 9. Generally MIG compression technique optimize the ratio and the JPEG compression quality in software. But MIG is longer

in time excecution process because of increasing insertion bits The future research is applicating develop method in for video compression.

References

[1] Image compression - from dct to wavelets : A review.

[2] The latest png specification.w3c tech reports.W3C Tech Reports. Oct. 1, 1996 is the recommended v 1.0 spec.

[3] Itu-t recommendation t.81: Information technology digital compression and coding of continuous-tone still images - requirements and guidelines, 1992.

[4] Iso/iec fcd 15444-1: Information technology :Jpeg2000 image coding system, part 1: Core coding system, 2004.

[5] C. Christopoulos A. Skodras and T. Ebrahimi.

The jpeg 2000 still image compression standard. In IEEE Signal Processing Magazine,

number 18, Sept 2001.

[6] E. Atsumi and N. Farvardin. \lossy/lossless region-of-interest image coding based on set partitioning in hierarchical trees. In IEEE InternationalConference on Image Processing, Oct. 1998.

[7] R.M. van Schyndel N.R.A. Mee C.F. A.Z.Tirkel,G.A.Rankin. Osborne. electronic water mark.In DICTA, 1993.

[8] Guy E. Blelloch. Computer science department,carnegie mellon university. Introduction to Data Compression, October 2001.

[9] A. P. Bradley and F. W. M. Stentiford. Jpeg2000 and region of interest coding. In Proc. Digital Image Computing Techniques and Applications, 2002.

[10] J. Askelof C. A. Christopoulos and M. Larsson. Efficient region of interest encoding techniques in the upcoming jpeg2000 still image coding standard,. In Proc. IEEE International Conference on Image Processing, Sept 2000.

[11] E. Ordentlich D. Taubman and I. Ueno. Embedded block coding. In Proc. IEEE International Conference on Image Processing, 2000.

[12] Alexey Moskvin Dmitriy Vatolin. JPEG 2000

Image Codecs Comparison, CS MSU Graphics&Media Lab Video Group. September 2005.

[13] P.E Douglas A. Kerr. JPEG Compression of Still

Images. August 16 2003.

[14] J. Zhao E. Koch. Towards robust and hidden imagecopyright labeling. In IEEE Workshop on

Non Linier Signal and Image Processing, June 1995.

[15] Adrian Ford and Alan Roberts. Colour space conversions, August 1998.

[16] et al G.C. Langelaar. Copy Protection for Multimedia Data based on Labeling Techniques.

1996.

[17] Bernd Girod. Ee368: Digital image processing, spring 2005-2006.

[18] Gernot Hoffmann. Jpeg compression,September 18 2003.

[19] http://www.acdsee.com/.

[20] http://www.morgan multimedia.com/.

[21] http://www.stanford.edu/ esetton/experiments2.htm.

[22] Ziv. J. and A. Lempel. A universal algorithm for sequential data compression. In IEEE Transactions on Information Theory, volume 23, pages 337–343, 1977.

[23] Ziv. J. and A. Lempel. Compression of individual sequences via variable-rate coding. In

IEEE Transactions on Information Theory, volume 24, September 1978.

[24] F.T. Leighton T. Shamoon J. Cox, J. Kilian. Secure spread spectrum watermarking for multimedia. In IEEE Transaction on Image Processing, December 1997.

[25] Sarifuddin Madenda. Multimedia data compression : today and future. Multimedia Information Research Group, 2006. Canada.

[26] R. Polika. The engineer’s ultimate guide to wavelet analysis: The wavelet tutoria.

[27] Zhigang Fan Sanjeeb Dash Ramesh Neelamani_, Ricardo de Queiroz and Richard G.

Baraniuk. Jpeg compression history estimation for color images. In IEEE Transactions on

Image Processing.

[28] R.E. Woods R.C. Gonzales. Digital Image Processing, Second Edition. Pearson Prentice Hall,

2005.

[29] B. Schneiner. Applied Cryptography: Protocols,Algorithm, and Source Code in C. Wiley,

New York, 1994.

[30] J. M. Shapiro. Embedded image coding using zerotrees of wavelet coeficients. In IEEE Transactions on Signal Processing, volume 41, Desc 1993.

[31] Arno Swart. An introduction to JPEG compression using MATLAB. October 2003.

[32] D. Taubman and M. W. Marcelin. Jpeg2000: Standard for interactive imaging. In Proceedings of the IEEE, volume 17.

[33] D. S. Taubman and M. W. Marcelin. JPEG 2000: Image Compression Fundamentals, Standards and Practice. Kluwer Academic Publishers, Dordrecht, 2002.

[34] N. Morimoto W. Bender, D. Gruhl. A. lu, techniquesfor data hiding. In IBM System Journal,

volume 35, 1996.

[35] G. K. Wallace. The JPEG still picture compression standard, volume 34. April, April 1991.

[36] Panrong Xiao. Image compression by wavelet transform, August 2001.

fajar uno