Image Compression using Digital Curvelet Transform and HWT as MCA
Navjot Kaur, Deepa Verma · International Journal of Computer Applications · 2013
Image compression has been always a very active field of research.A highly efficient numerical scheme is proposed to solve the combined optimization problem posed by the model for separating images into texture and piecewise smooth parts.In the proposed multi-layered image coding schemes, the MCA used in image decomposition is performed using haar wavelet transform that decomposes the image into four frequency sub-band.The results show that the proposed algorithm that is the combination of wavelet based decomposition as extraction of texture and edge parts using the haar wavelet transform and further compressing of texture and edge part using dct and the Curvelet transform respectively, give the enhanced PSNR and other statistical parameters.The results are evaluated in different bits per pixels (bpp) color format and are in a proportionate order.i.e. as the bpp increases, the PSNR improves.Other image compression performance parameters like Standard Deviation, Entropy, Compression Ratio and Class Variance are evaluated to analyse the compression performance.