A novel method for optimizing the ratio of image compression

In this paper, a method for increasing the compression ratio, is presented. This method is sensitive to the smoothness of each block in the image, where, the criterion of smoothness is the standard deviation. For a certain level of smoothness, the block can be approximated and represented by only its spectral DC component. The resulted blocking artefacts, introduced during this compression process, can be enhanced by smoothing of the blocks boundaries. Two factors are affecting the compression ratio, The first is the smoothing factor (threshold) for standard deviation. The second is the used quantization table. Experimental results, using MATLAB, demonstrate that the proposed algorithm can optimize the trade-off between compression ratio and quality of image.