Keywords : lightness logarithmic


Digital Image Contract Computation and Enhancement Using Statistical Methods

Abdullah Hammod Muhammad

Kirkuk University Journal-Scientific Studies, 2013, Volume 8, Issue 1, Pages 1-24
DOI: 10.32894/kujss.2013.77595

The optical imaging systems, do not give a perfect and ideal image. But mostly give images with limited resolutions i.e. give images with blur edges. This blurring attributed to two main reasons. The diffraction of light and Aberrations of the optical imaging system. The most important information and details in the image that contained in image edges. The fidelity criteria are evaluate the quality, depending on the whole image plane or in homogenous image regions. So that we devoted in this study to evaluate image quality depending on compute the image contrast in edge regions, and introduce robust quantitative measures to determine image quality, then estimate the efficiency of the various techniques in image processing applications. In this study we suggested new techniques to calculate image contrast (visibility) and studying it as a function of number of smoothing iterations from using mean filter and a function of gray level resolution. We only study the contrast in edge regions where we used Soble operator to find image edges. The suggested techniques are:-
1- The direct technique for compute image contrast this depending on determine the largest, and smallest image elements in edge regions.
2- The statistical method to estimate the contrast that based on determines the mean and standard deviation in the image edge regions.
The results give high agreement among the various suggested methods in determines image contrast. As we can theoretical guested, that the contrast reduced with reducing spatial and image gray level resolutions.
In this study we also apply the direct and statistical methods to evaluate the performance of histogram equalization and the logarithmic illumination enhancement techniques. Where the results, show that the contrast enhanced by using histogram equalization and reduced by using logarithmic illumination enhancement method. Here can be say that we get a robust quantitative measures that could be used to estimate the efficiency of the image processing techniques, based on determine image contrast and find the amount of variation in contrast that causes from the processing steps.