Multi-class Image Coding via EM-KLT Algorithm

This paper presents a new image coding method in which the image blocks are assigned to different classes learned by the EM algorithm. Each class is matched to a multidimensional Gaussian density function and the Karhunen-Loeve Transform (KLT), followed by optimal quantization and coding, is applied to each one of them. The performance of this Class-KLT coder is compared to the classical KLT coder (one class) showing appreciable improvement in image quality.

[1]  Bhaskar D. Rao,et al.  PDF optimized parametric vector quantization of speech line spectral frequencies , 2003, IEEE Trans. Speech Audio Process..

[2]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[3]  Anil K. Jain.  Fundamentals of Digital Image Processing , 2018, Control of Color Imaging Systems.

[4]  Antonio Ortega,et al.  Adaptive scalar quantization without side information , 1997, IEEE Trans. Image Process..

[5]  David G. Stork,et al.  Pattern Classification , 1973 .

[6]  David G. Stork,et al.  Pattern Classification (2nd ed.) , 1999 .

[7]  Allen Gersho,et al.  Vector quantization and signal compression , 1991, The Kluwer international series in engineering and computer science.

[8]  Robert M. Gray.  Gauss mixture vector quantization , 2001, 2001 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings (Cat. No.01CH37221).