BTC Image Coding Using Vector Quantization
- 1 March 1987
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Communications
- Vol. 35 (3) , 352-356
- https://doi.org/10.1109/tcom.1987.1096773
Abstract
This paper describes source encoding of the outputs of a block truncation coder (BTC), namely, the overhead statistical information and the truncated block. The statistical overhead and the truncated block exhibit properties which can be effectively used for their quantization as vectors. Vector quantization of these BTC outputs results into reduction of the bit rate of the coder. The bit rate reduces up to 1.5 bits/ pel if vector quantization is used on one of the outputs; i.e., either the overhead information or the truncated block. By vector quantizing both the BTC outputs the bit rate can he reduced up to 1.0 bits/pel without introducing many perceivable errors in the reconstructed output.Keywords
This publication has 8 references indexed in Scilit:
- Modified algorithm for block truncation coding of monochrome imagesElectronics Letters, 1985
- A generalized block truncation coding algorithm for image compressionIEEE Transactions on Acoustics, Speech, and Signal Processing, 1984
- Vector quantizationIEEE ASSP Magazine, 1984
- BTC Image Coding Using Median Filter RootsIEEE Transactions on Communications, 1983
- Vector quantiser of video signalsElectronics Letters, 1982
- Digital Video Bandwidth Compression Using Block Truncation CodingIEEE Transactions on Communications, 1981
- An Algorithm for Vector Quantizer DesignIEEE Transactions on Communications, 1980
- Image Compression Using Block Truncation CodingIEEE Transactions on Communications, 1979