Dual-tree complex wavelet with modified SPIHT algorithm for video compression

Abstract

For efficient video compression, the 3 Dimensional Set-Partitioning in Hierarchical Trees (3D-SPIHT) algorithm and 3D wavelet transform are used. The 3D wavelet transform is based on the group of frames concept. For decomposition, wavelet filters are used. The transform coefficients are coded by 3D SPIHT. In the reconstruction phase, 3D SPIHT decoding algorithm and inverse transforms are employed. A modified SPIHT algorithm called Block Based Pass Parallel SPIHT (BPSPIHT) is employed for video compression to overcome the slow processing speed of SPIHT. The video is detached into individual frames. Then Discrete Wavelet Transform (DWT) and Dual Tree Complex Wavelet Transform (DTCWT) are employed on that frames. The wavelet coefficients are encoded using BPSIHT encoder and the generated bit streams are decoded by BPSPIHT decoder. For the reconstruction, inverse wavelet transform is applied and the calibre is measured in terms of PSNR, MSE and Compression Ratio (CR).