Statistical model of quantized DCT coefficients

Author

Zou, Fuliao ; Lu, Zhengding ; Hefei Ling

Author_Institution

Sch. of Comput. Sci. & Technol., Huazhong Univ. of Sci. & Technol., Wuhan, China

Volume

3

fYear

2004

Firstpage

2572

Abstract

In this paper we propose a two-parameter probability model of quantized DCT coefficients for image and video (i.e. JPEG&MPEG format materials), which is based on empirical observation of their statistics in the quantized DCT coefficient domain. The parameters are evaluated in terms of the maximum likelihood criterion which is applied to different frequency coefficients space. The experimental results show that quantized DCT coefficients´ distribution modeled by generalized Laplacian performs better than by Laplacian.

Keywords

Laplace transforms; data compression; discrete cosine transforms; maximum likelihood estimation; probability; quantisation (signal); video coding; digital watermarking; discrete cosine transform; generalized Laplacian distribution; image; maximum likelihood criterion; probability density function; quantized DCT coefficient; video; Discrete cosine transforms; Discrete wavelet transforms; Frequency; Image coding; Laplace equations; Maximum likelihood estimation; Probability density function; Steganography; Video compression; Watermarking;

fLanguage

English

Publisher

ieee

Conference_Titel

Signal Processing, 2004. Proceedings. ICSP '04. 2004 7th International Conference on

Print_ISBN

0-7803-8406-7

Type

conf

DOI

10.1109/ICOSP.2004.1442307

Filename

1442307