WAVELET FILTERS FOR IMAGE COMPRESSION, AN ANALYTICAL STUDY

JPEG 2000 is the current still image compression standard which adopts a wavelet based technology in its compression scheme. The first step in the algorithm is to wavelet transform. This paper presents a complete analytical study and characterization of wavelet filters used in wavelet based image coding. Most commonly used image compression algorithm Set Partitioning in Hierarchical Trees (SPIHT) is used to compare the different wavelet functions. The study includes the effect of different images, level of decomposition, and bitrates, bits per pixel (bpp). Simulation results show the important properties of wavelet have to be considered for image compression. Peak signal to noise ratio (PSNR) is used as a measure to compare wavelet filters.