Data compression using wavelet transforms and vector quantization

In this paper a general purpose compression coding algorithm is described, applicable to both sound and image data. Using recent developments in Discrete Wavelet Transforms (DWT) and Vector Quantization (VQ) the algorithm is compared to Direct Vector Quantization (DVQT) methods. Test results show a 19% to 84% improvement in SNR over DVQT for speech data and a 17% to 50% improvement in SNR for image data.