Volume data compression using model-based quantization and three-dimensional zerotree encoding of wavelet coefficients

Volume data such as those acquired by magnetic resonance imaging (MRI) techniques can be compressed efficiently using wavelet transform-based methods. This approach uses a three-dimensional wavelet transform to reduce the spatial redundancies in the volume data. A model-based quantization scheme is used to quantize the wavelet coefficients. Unlike methods based on other transforms, wavelet compression methods need to retain both the value and the location of the significant coefficients. A three-dimensional zerotree encoding method is used to efficiently encode the quantized symbols. In addition to quantitative measures, an assessment is performed of the compression results using a volume visualization process.