This paper suggests an algorithm of multiresolutional Classified Vector Quantization for Image Coding

in according to the correlation among the various scales and different orientation bands after wavelet transformation.This algorithm rearranges the coefficients from various resolution level and orientation and gets a vector of 64 dimensions which then is classified and quantized.Meantime

we design an efficient classifier

and give the experiment results and data under different conditions.Compaered with other methods

it proved that the algorithm not only improves the image quality

but also decreases the computation complexity.