Title :
Optimal quantizer step sizes for transform coders
Author :
Hung, Andy C. ; Meng, Teresa H -Y
Author_Institution :
Comput. Syst. Lab., Stanford Univ., CA, USA
Abstract :
As evidenced by the theoretical modeling and simulation presented, the optimal quantizer solution for run-length and Huffman coding is quite addressable for the Laplacian case. The step sizes may be calculated in advance for images that share common statistics. Furthermore, the model can also be used for other uses such as bit-management for rate control and lower bound estimation for compressed file sizes; it is also the first step in the general solution to the transform coder quantization problem
Keywords :
encoding; picture processing; transforms; Huffman coding; JPEG algorithm; Laplacian case; bit-management; compressed file sizes; images; lower bound estimation; modeling; optimal quantizer step sizes; rate control; run length coding; simulation; statistics; transform coder quantization; Code standards; Discrete cosine transforms; Discrete transforms; Frequency; Laboratories; Laplace equations; Pulse modulation; Quantization; Statistics; Transform coding;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1991. ICASSP-91., 1991 International Conference on
Conference_Location :
Toronto, Ont.
Print_ISBN :
0-7803-0003-3
DOI :
10.1109/ICASSP.1991.150939