Image compression system incorporating wavelets, vector quantization, and adaptive filtering

Date of Award




Degree Name

Doctor of Philosophy (Ph.D.)


Electrical and Computer Engineering

First Committee Member

Claude S. Lindquist, Committee Chair


The wavelet transform (WT) decomposes the image into a set of subband images with different spatial resolutions corresponding to different frequency bands. The WT coding provides superior image quality at low bit rates since it minimizes the blocking effects otherwise obtained when using the discrete cosine transform (DCT). In a subband coding system in addition to the WT on the image there is quantization, hence information loss occurs manifesting itself in signal dependent quantization noise.A new method for dealing with the effects of vector quantization (VQ) in a subband system is proposed. It uses Lloyd-Max quantizers in a two stage quantization scheme. It also uses an optimal linear prediction to decorrelate the quantization noise and reduce the non-linear dependency on the signal. Thereafter a local adaptive orientation Wiener (AOW) filter is applied to reduce the quantization noise and preserve the image edges. An important feature of the new method is the reduction in computational complexity of the AOW filter and the vector quantizer which is achieved due to the reduction in signal dependency of the noise and shorter vector lengths.


Engineering, Biomedical; Engineering, Electronics and Electrical

Link to Full Text


Link to Full Text