DefinePK

Abstract

This paper will address a new proposed algorithm using wavelet properties to compress an image. It is based on the concept of reducing the near-zero wavelet coefficients at detail subbands (Diagonal, Vertical and Horizontal). This approach is inspired by formerly known Hard Thresholding that eliminate all the coefficient under the fixed threshold value while keeping up the rest. Here, we proposed our own threshold value estimation based on standard deviation concept to find the optimal threshold value at each details subbands. Throughout the experiment done, we found that the proposed algorithm can effectively remove a large amount of unnecessary wavelet coefficient without harming the image quality while increasing the compression ratio and reducing the elapse time.