| | |
Summary: Wavelet Based Subband Vector Quantization Algorithm for Gray Images
Tansu Alpcan, Mustafa Kesal and Hakan Deliç
Signal and Image Processing Laboratory (BUSIM), Department of Electrical and Electronics
¢¡¤£¤¥¦¡¨§©§¥¦¡¨£"!©¥$#©¥¨%&¡'¥ ('§0)1¥¦24356§7¤§8@9BA¨9DC"EGFH)H2IP¡¤7Q'R$
Turkey
Abstract- Wavelet decomposition coefficients
of a gray level image are vector quantized
with flexible, variable sized codewords and
a wavelet based tree search algorithm.
Vector quantization codewords are designed
as rectangles with ratio 1 to 2, which is more
efficient in encoding different subbands
with minimum error. The performance of
the discrete cosine transform is observed in
different bands. The low frequency subband
is scalar quantized due to the poor
performance of DCT in this band.
I. INTRODUCTION
Subband decomposition is a widely used
and accepted technique for image
|
