Top Banner
Base paper: - http://www.ijettjournal.org/volume-4/issue-4/IJETT-V4I4P263.pdf RGB Image Compression Using Two Dimensional Discrete Cosine Transform International Journal of Engineering Trends and Technology Abstract: Recent research in transform-based image compression has focused on the wavelet transform due to its superior performance over other transforms. Performance is often measured solely in terms of peak signal-to-noise ratio (PSNR) and compression algorithms are optimized for this quantitative metric. The performance in terms of subjective quality is typically not evaluated. Moreover, the sensitivities of the human visual system (HVS) are often not incorporated into compression schemes. This thesis develops new wavelet models of the HVS and illustrates their performance for various scalar wavelets and multi-wavelet transforms. The performance is measured quantitatively (PSNR) and qualitatively using our new perceptual testing procedure. Our new HVS model is comprised of two components: CSF masking and asymmetric compression. CSF masking weights the wavelet coefficients according to the contrast sensitivity function (CSF)–a model of humans’ sensitivity to spatial frequency. This mask gives the most perceptible information the highest priority in the quantizer. The second component of our HVS model is called asymmetric compression. It is well known that humans are more sensitive to luminance stimuli than they are to chrominance stimuli; asymmetric compression quantizes the chrominance spaces more severely than the luminance component. The results of extensive trials indicate that our HVS model improves both quantitative and qualitative performance. These trials included 14 observers, 4 gray-scale images and 10 color images (both natural and synthetic). For gray-scale images, although our HVS scheme lowers PSNR, it improves subjective quality. For color images, our HVS model improves both PSNR and subjective quality. A benchmark for our HVS method is the latest version of the international image compression standard–JPEG2000. In terms of subjective quality, our scheme is superior to JPEG2000 for all images; it also outperforms JPEG2000 by 1 to 3 dB in PSNR.
11
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: RGB Image Compression using Two-dimensional Discrete Cosine Transform

Base paper: - http://www.ijettjournal.org/volume-4/issue-4/IJETT-V4I4P263.pdf

RGB Image Compression Using Two Dimensional Discrete Cosine Transform

International Journal of Engineering Trends and Technology

Abstract:

Recent research in transform-based image compression has focused on the wavelet transform due to its superior performance over other transforms. Performance is often measured solely in terms of peak signal-to-noise ratio (PSNR) and compression algorithms are optimized for this quantitative metric. The performance in terms of subjective quality is typically not evaluated. Moreover, the sensitivities of the human visual system (HVS) are often not incorporated into compression schemes. This thesis develops new wavelet models of the HVS and illustrates their performance for various scalar wavelets and multi-wavelet transforms. The performance is measured quantitatively (PSNR) and qualitatively using our new perceptual testing procedure. Our new HVS model is comprised of two components: CSF masking and asymmetric compression. CSF masking weights the wavelet coefficients according to the contrast sensitivity function (CSF)–a model of humans’ sensitivity to spatial frequency. This mask gives the most perceptible information the highest priority in the quantizer. The second component of our HVS model is called asymmetric compression. It is well known that humans are more sensitive to luminance stimuli than they are to chrominance stimuli; asymmetric compression quantizes the chrominance spaces more severely than the luminance component. The results of extensive trials indicate that our HVS model improves both quantitative and qualitative performance. These trials included 14 observers, 4 gray-scale images and 10 color images (both natural and synthetic). For gray-scale images, although our HVS scheme lowers PSNR, it improves subjective quality. For color images, our HVS model improves both PSNR and subjective quality. A benchmark for our HVS method is the latest version of the international image compression standard–JPEG2000. In terms of subjective quality, our scheme is superior to JPEG2000 for all images; it also outperforms JPEG2000 by 1 to 3 dB in PSNR.

Page 2: RGB Image Compression using Two-dimensional Discrete Cosine Transform

Base paper: - http://www.ijettjournal.org/volume-4/issue-4/IJETT-V4I4P263.pdf

(a) Block Diagram

(b) Block diagram of the Color Image Compression Algorithm

(c) The Analysis and Synthesis Stages of a 2-D, 1 Level Scalar Wavelet Decomposition

Page 3: RGB Image Compression using Two-dimensional Discrete Cosine Transform

Base paper: - http://www.ijettjournal.org/volume-4/issue-4/IJETT-V4I4P263.pdf

(d) Comparison of One-Level (a) Scalar Wavelet, and (b) Multi-wavelet Decomposition

(e) The Analysis Stage of a 2-D, 1-Level Multi-wavelet decomposition with r = 2

Page 4: RGB Image Compression using Two-dimensional Discrete Cosine Transform

Base paper: - http://www.ijettjournal.org/volume-4/issue-4/IJETT-V4I4P263.pdf

(f) Block Diagram of CSF Masking Method

Page 5: RGB Image Compression using Two-dimensional Discrete Cosine Transform

Base paper: - http://www.ijettjournal.org/volume-4/issue-4/IJETT-V4I4P263.pdf

A: -

Page 6: RGB Image Compression using Two-dimensional Discrete Cosine Transform

Base paper: - http://www.ijettjournal.org/volume-4/issue-4/IJETT-V4I4P263.pdf

B: -

Page 7: RGB Image Compression using Two-dimensional Discrete Cosine Transform

Base paper: - http://www.ijettjournal.org/volume-4/issue-4/IJETT-V4I4P263.pdf

C: -

Page 8: RGB Image Compression using Two-dimensional Discrete Cosine Transform

Base paper: - http://www.ijettjournal.org/volume-4/issue-4/IJETT-V4I4P263.pdf

D: -

Page 9: RGB Image Compression using Two-dimensional Discrete Cosine Transform

Base paper: - http://www.ijettjournal.org/volume-4/issue-4/IJETT-V4I4P263.pdf

E: -

Page 10: RGB Image Compression using Two-dimensional Discrete Cosine Transform

Base paper: - http://www.ijettjournal.org/volume-4/issue-4/IJETT-V4I4P263.pdf

Conclusion and Future Enhancements:

This thesis provides the most comprehensive results to date for wavelet-based image compression with a consideration for the HVS. Although our results indicate better quantitative and qualitative performance than previous research, avenues for improvement remain. 1. The CSF masking operation presented in this thesis applies to scalar wavelets and multi-wavelets. We believe that the same idea can be extended to wavelet packets and multiwavelet packets. However, constructing a CSF mask for wavelet packets and multiwavelet packets is not trivial. We anticipate that constructing and applying a CSF mask to a wavelet packet or multiwavelet packet decomposition will provide quantitative and qualitative improvement similar to what we have seen for scalar wavelets and multi-wavelets. 2. Our color compression scheme transforms the RGB input image into the YCbCr color space. This is a standard method–it is used in JPEG. However, other color space models exist and may offer a more suitable breakdown for wavelet-based compression. Among the other options for color compression is the opponent color space model that has been used in recent compression research [16]. 3. The same HVS compression ideas used here for still images may provide similar improvement for digital video compression. CSF masking is a simple operation that can be applied on a frame by frame basis while adding minimal computational overhead. 4. Both quantitatively and qualitatively, the best multiwavelet rarely outperforms the best scalar wavelet. New multi-filters are continuing to be developed. The family of balanced multi-wavelets [33] and symmetric FIR balanced multi-wavelets [21] may provide performance superior to the multi-wavelets used in this thesis. 5. CSF masking was not performed in the two chrominance spaces because humans’ sensitivity to chrominance stimuli is relatively uniform across frequency. However, potential gain may be found in changing the wavelet method in each color space. For instance, multi-wavelets may potentially compress chrominance information better than scalar wavelets. If trends can be found, we can choose to compress each color space with a particular wavelet method. Such a scheme may provide additional quantitative and/or qualitative gain.

Page 11: RGB Image Compression using Two-dimensional Discrete Cosine Transform

Base paper: - http://www.ijettjournal.org/volume-4/issue-4/IJETT-V4I4P263.pdf

References: [1] R. C. Gonzalez, R. E. Woods, “Digital image processing,” Pearson Education, III Edition, 2008 [2] David Salomon, “Data compression” springer, Fourth Edition, 2007 [3] Jiankun Li, Jin Li, and C.-C. Jay Kuo, “Layered DCT Still Image Compression”, IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, VOL. 7, NO. 2, pp. 440-443, APRIL 1997 [4] Zixiang Xiong, Kannan Ramchandran, Michael T. Orchard, and Ya-Qin Zhang,” A Comparative Study of DCT- and Wavelet-Based Image Coding”, IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, VOL. 9, NO. 5, pp. 692- 695, AUGUST 1999 [5] Panos Nasiopoulos and Rabab K. Ward, “A High-Quality Fixed- Length Compression Scheme for Color Images”, IEEE TRANSACTIONS ON COMMUNICATIONS, VOL. 43, NO. 11, pp. 2672-2677, NOVEMBER 1995 [6] K. R. Rao and P. Yip, Discrete Cosine Transform. New York: Academic, 1990. [7] J. M. Shapiro, “Embedded image coding using zero trees of wavelet coefficients,” IEEE Trans. Signal Processing, vol. 41, pp. 3445– 3463, Dec. 1993. [8] A. Said and W. A. Pearlman, “A new, fast, and efficient image codec based on set partitioning in hierarchical trees,” IEEE Trans. Circuits Syst. Video Technol., vol. 6, pp. 243–250, June 1996. [9] D.Wu and E.C.Tan,“Comparison of loss less image compression”. TENCON 99 proceedings of the IEEE region 10 conference, Cheju island, South korea, vol.1, 1999, pages 718-721 [10] Jagadish H. Pujar and Lohit M. Kadlaskar, “A NEW LOSSLESS METHOD OF IMAGE COMPRESSION AND DECOMPRESSION USING HUFFMAN CODING TECHNIQUES” JATIT, Edition, 2005-2010 [11] Lina J. Karam, “Handbook of image and video processing” Second Edition, 2005, Pages 643-660 [12] M. H. El Ayadi M. M. Syiam A. A. Gamgoum” A Comparative study of lossless image compression techniques” IJICIS, Vol. 5, No. 1, July 2005 [13] Chaur-Chin Chen, "On the selection of image compression algorithms," 14th International conference on pattern recognition (ICPR'98) - Volume 2, 1998. [14] Amhamed saffor, Abdul Rahman Ramli, Kwan-Hoong Ng, “A Comparative study of Image Compression between JPEG and WAVELET”. Malaysian journal of computer science, vol. 14 no. 1, June 2001, Pages.39-45. [15] Mr. T. Sreenivasulu reddy, Ms. K. Ramani, Dr. S. Varadarajan, Dr. B.C.Jinaga “Image compression using transform coding methods” , International journal of Computer science and network security, vol.7 no.7, July 2007. [16] Andrew B. Watson” Image compression using the Discrete Cosine Transform” Mathematica journal, 4(1), 1994, Pages 81-88