Wavelet Coefficients Reduction Method Based on Standard Deviation Concept for High Quality Compressed Image

This paper proposes a method to estimate the wavelet coefficient threshold value at wavelet details subbands to archive high quality image compression. It uses the features of Discrete Wavelet Transform (DWT) where the images are decomposed into one approximate subband and three detail subbands. The detail subbands consists of Diagonal, Vertical and Horizontal subband. By using the standard deviation concept we estimate the threshold value at each detail subbands. The coefficients with larger value then the calculated threshold will be retained while the rest will be discarded. So, for every single image, different threshold values are calculated and applied at each detail subbands. From the experiment done, we found that the proposed method eliminate the near-zero coefficients effectively without harming the quality of the compressed image.