Vector quantization for multiple classes
- 1 January 1997
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 4, 93-102 vol.4
- https://doi.org/10.1109/aero.1997.577500
Abstract
Vector quantization algorithms have long been used to find a finite set of exemplars which represent a data set to within an a priori error tolerance. Such a representation is essential in codebook-based data compression and transmission. The present study considers the situation where the data to be encoded consists of subclasses. The codebook must provide information compression within the several subclasses; however, minimization of interclass errors is of equal importance. We present modifications to a basic vector quantization (VQ) algorithm which adapts it to the multiclass vector quantizing setting. We then explore the behavior of the modified algorithm on selected benchmark applications. We show, in particular, that overlapping subclasses can be accommodated by the algorithm.Keywords
This publication has 10 references indexed in Scilit:
- Binary search trees for vector quantisationPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2005
- VQ assistance for training perceptron networksPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Efficient exemplars for classifier designJournal of the Franklin Institute, 1995
- A new vector quantization clustering algorithmIEEE Transactions on Acoustics, Speech, and Signal Processing, 1989
- Vector quantizationIEEE ASSP Magazine, 1984
- Least squares quantization in PCMIEEE Transactions on Information Theory, 1982
- On the structure of vector quantizersIEEE Transactions on Information Theory, 1982
- An Algorithm for Vector Quantizer DesignIEEE Transactions on Communications, 1980
- An Algorithm for Finding Best Matches in Logarithmic Expected TimeACM Transactions on Mathematical Software, 1977
- Multidimensional binary search trees used for associative searchingCommunications of the ACM, 1975