Title :
Analysis of the DCT coefficient distributions for document coding
Author_Institution :
Dept. of Electr. & Electron. Eng., Univ. of Hong Kong, China
Abstract :
It is known that the distribution of the discrete cosine transform (DCT) coefficients of most natural images follow a Laplacian distribution, and this knowledge has been employed to improve decoder design. However, such is not the case for text documents. In this letter, we present an analysis of their DCT coefficient distributions, and show that a Gaussian distribution can be a realistic model. Furthermore, we can use a generalized Gaussian model to incorporate the Laplacian distribution found for natural images.
Keywords :
Gaussian distribution; Laplace transforms; discrete cosine transforms; document image processing; image coding; transform coding; DCT coefficient distribution analysis; Laplacian distribution; decoder design; discrete cosine transform; document coding; image coding; natural image analysis; probability statistics; realistic Gaussian distribution model; text document processing; Discrete cosine transforms; Gaussian distribution; Gray-scale; Image analysis; Image coding; Laplace equations; Maximum likelihood decoding; Stochastic processes; Text analysis; Transform coding;
Journal_Title :
Signal Processing Letters, IEEE
DOI :
10.1109/LSP.2003.821789
