A multi-objective evolutionary approach to image quality/compression trade-off in JPEG baseline algorithm

The JPEG algorithm is one of the most used tools for compressing images. The main factor affecting the performance of the JPEG compression is the quantization process, which exploits the values contained in two tables, called quantization tables. The compression ratio and the quality of the decoded images are determined by these values. Thus, the correct choice of the quantization tables is crucial to the performance of the JPEG algorithm. In this paper, a two-objective evolutionary algorithm is applied to generate a family of optimal quantization tables which produce different trade-offs between image compression and quality. Compression is measured in terms of difference in percentage between the sizes of the original and compressed images, whereas quality is computed as mean squared error between the reconstructed and the original images. We discuss the application of the proposed approach to well-known benchmark images and show how the quantization tables determined by our method improve the performance of the JPEG algorithm with respect to the default tables suggested in Annex K of the JPEG standard.

[1]  Gregory K. Wallace,et al.  The JPEG still picture compression standard , 1992 .

[2]  David W. Corne,et al.  Approximating the Nondominated Front Using the Pareto Archived Evolution Strategy , 2000, Evolutionary Computation.

[3]  Teresa H. Meng,et al.  Optimal quantizer step sizes for transform coders , 1991, [Proceedings] ICASSP 91: 1991 International Conference on Acoustics, Speech, and Signal Processing.

[4]  N. Ahmed,et al.  Discrete Cosine Transform , 1996 .

[5]  Mandyam D. Srinath,et al.  Statistical distributions of image DCT coefficients , 1986 .

[6]  K A Birney,et al.  On the modeling of DCT and subband image data for compression , 1995, IEEE Trans. Image Process..

[7]  F. Muller.  Distribution shape of two-dimensional DCT coefficients of natural images , 1993 .

[8]  Edmund Y. Lam.  Analysis of the DCT coefficient distributions for document coding , 2004, IEEE Signal Processing Letters.

[9]  Barry G. Sherlock,et al.  Optimum DCT quantization , 1993, [Proceedings] DCC `93: Data Compression Conference.

[10]  Zbigniew Michalewicz,et al.  Genetic Algorithms + Data Structures = Evolution Programs , 1996, Springer Berlin Heidelberg.

[11]  Marco Laumanns,et al.  SPEA2: Improving the Strength Pareto Evolutionary Algorithm For Multiobjective Optimization , 2002 .

[12]  A.C.P. Veiga,et al.  Identification of the best quantization table using genetic algorithms , 2005, PACRIM. 2005 IEEE Pacific Rim Conference on Communications, Computers and signal Processing, 2005..

[13]  Lothar Thiele,et al.  Multiobjective evolutionary algorithms: a comparative case study and the strength Pareto approach , 1999, IEEE Trans. Evol. Comput..

[14]  David E. Goldberg,et al.  A niched Pareto genetic algorithm for multiobjective optimization , 1994, Proceedings of the First IEEE Conference on Evolutionary Computation. IEEE World Congress on Computational Intelligence.

[15]  Zbigniew Michalewicz,et al.  Genetic algorithms + data structures = evolution programs (2nd, extended ed.) , 1994 .

[16]  Hocine Cherifi,et al.  On the distribution of the DCT coefficients , 1994, Proceedings of ICASSP '94. IEEE International Conference on Acoustics, Speech and Signal Processing.

[17]  Kalyanmoy Deb,et al.  A fast and elitist multiobjective genetic algorithm: NSGA-II , 2002, IEEE Trans. Evol. Comput..

[18]  Trac D. Tran,et al.  A locally adaptive perceptual masking threshold model for image coding , 1996, 1996 IEEE International Conference on Acoustics, Speech, and Signal Processing Conference Proceedings.

[19]  D. M. Monro,et al.  A model for JPEG quantization , 1994, Proceedings of ICSIPNN '94. International Conference on Speech, Image Processing and Neural Networks.

[20]  Shen-Chuan Tai,et al.  An adaptive 3-D discrete cosine transform coder for medical image compression , 2000, IEEE Transactions on Information Technology in Biomedicine.

[21]  Giuliano Benelli,et al.  A DCT-based adaptive compression algorithm customized for radar imagery , 1997, IGARSS'97. 1997 IEEE International Geoscience and Remote Sensing Symposium Proceedings. Remote Sensing - A Scientific Vision for Sustainable Development.

[22]  Joseph W. Goodman,et al.  A mathematical analysis of the DCT coefficient distributions for images , 2000, IEEE Trans. Image Process..

[23]  Kalyanmoy Deb,et al.  Muiltiobjective Optimization Using Nondominated Sorting in Genetic Algorithms , 1994, Evolutionary Computation.

[24]  Lothar Thiele,et al.  Comparison of Multiobjective Evolutionary Algorithms: Empirical Results , 2000, Evolutionary Computation.

[25]  K. R. Rao,et al.  Techniques and Standards for Image, Video, and Audio Coding , 1996 .

[26]  Jerry D. Gibson,et al.  Distributions of the Two-Dimensional DCT Coefficients for Images , 1983, IEEE Trans. Commun..

[27]  Joan L. Mitchell,et al.  JPEG: Still Image Data Compression Standard , 1992 .

[28]  Y.-G. Wu,et al.  GA-based DCT quantisation table design procedure for medical images , 2004 .

[29]  S. Liu,et al.  Statistical analysis of the DCT coefficients and their quantization error , 1996, Conference Record of The Thirtieth Asilomar Conference on Signals, Systems and Computers.