An adaptive perception-based image preprocessing method

The aim of this paper is to introduce an adaptive preprocessing procedure based on human perception in order to increase the performance of some standard image processing techniques. Specifically, image frequency content has been weighted by the corresponding value of the contrast sensitivity function, in agreement with the sensitiveness of human eye to the different image frequencies and contrasts. The 2D Rational dilation wavelet transform has been employed for representing image frequencies. In fact, it provides an adaptive and flexible multiresolution framework, enabling an easy and straightforward adaptation to the image frequency content. Preliminary experimental results show that the proposed preprocessing allows us to increase the performance of some standard image enhancement algorithms in terms of visual quality and often also in terms of PSNR.