Image Compression Using KLT, Wavelets and an Adaptive Mixture of Principal Components Model
- 238 Downloads
In this paper, we present preliminary results comparing the nature of the errors introduced by the mixture of principal components (MPC) model with a wavelet transform and the Karhunen Loève transform (KLT) for the lossy compression of brain magnetic resonance (MR) images. MPC, wavelets and KLT were applied to image blocks in a block transform coding scheme. The MPC model partitions the space of image blocks into a set of disjoint classes and computes a separate KLT for each class. In our experiments, though both the wavelet transform and KLT obtained a higher peak signal to noise ratio (PSNR) than MPC, according to radiologists, MPC preserved the texture and boundaries of gray and white matter better than the wavelet transform or KLT.
KeywordsBrain Magnetic Resonance Image Image Compression Vector Quantization Image Block Lossy Compression
Unable to display preview. Download preview PDF.
- 4.J.N. Bradley, C.M. Brislawn, and T. Hopper, “The FBI wavelet/scalar quantization standard for gray-scale image compression”, in Visual Information Processing II, Vol. 1961 of SPIE Proceedings, Orlando, FL, pp. 293–304, April 1993.Google Scholar
- 7.A. Cohen, I.C. Daubechies, and J.C. Feauveau, “Biorthogonal bases of compactly supported wavelets”, Technical Report 11217-900529-07 TM, AT&T Bell Labs, Murray Hill, NJ, May 1990.Google Scholar
- 8.“WSQ gray-scale fingerprint image compression specification”, Technical report, Criminal Justice Information Services, Federal Bureau of Investigation, USA, Feb. 1993.Google Scholar
- 10.R.D. Dony and S. Haykin, “Optimally integrated adaptive learning”, Proceedings of the 1993 IEEE International Conference on Acoustics, Speech, and Signal Processing, IEEE, pp. I 609–611, April 1993.Google Scholar
- 12.R.D. Dony, “Adaptive transform coding of images using a mixture of principal components”, Ph.D. Thesis, McMaster University, Hamilton, Ontario, Canada, July 1995.Google Scholar
- 15.N. Kambhatla, Local models and Gaussian mixture models for statistical data processing, Ph.D. Thesis, Oregon Graduate Institute of Science & Technology, P.O. Box 91000, Portland, OR 97291-1000, USA, Jan. 1996.Google Scholar
- 16.N. Kambhatla and T.K. Leen, “Fast non-linear dimension reduction”, in Advances in Neural Information Processing Systems 6, Cowan, Tesauro, and Alspector (eds.), Morgan Kaufmann, San Mateo, California, pp. 152–159, 1994.Google Scholar
- 17.J. Domaszewicz and V.A. Vaishampayan, “Structural limitations of self-affine and partially self-affine fractal compression”, Proc. SPIE's Visual Communications and Image Processing, Cambridge, MA, Nov. 1993.Google Scholar
- 19.R.D. Dony, Adaptive transform coding of images using a mixture of principal components, Ph.D. Thesis, McMaster University, 1280 Main Street West, Hamilton, ON L8S4K1, Canada, July 1995.Google Scholar
- 21.R.D. Dony and S. Haykin, “Compression of SAR images using KLT, VQ, and mixture of principal components”, accepted to IEE Proceedings—Radar, Sonar and Navigation, 1997.Google Scholar
- 24.R.D. Dony, “Image compression using adaptive lapped transforms”, accepted to Canadian Conference on Electrical and Computer Engineering, 1997.Google Scholar