Enhancement of Chest X-ray Images for Diagnosis Purposes

Abstract

This study presents image quality comparison between original images and three image enhancement techniques namely imadjust, histogram equalization (HE) and contrast limited adaptive histogram equalization (CLAHE). These techniques are applied to a collection of eight chest x-ray images which are considered as dark, noisy and low in contrast and radiation dosage. The lacks of quality can be solved with these enhancement image techniques. These techniques raised the quality of images and improve the diagnostic ability of the pathological features observed in the images. Then the quality image factors including peak signal-to-noise ratio (PSNR), Mean squared error (MSE), (MAXERR) and (L2rat) were used to evaluate the characteristic of the images. The findings showed that the enhancement techniques managed to enhance the images make them more qualified than original images.

Keywords: Histogram Equalization, Contrast Limited Adaptive Histogram Equalization and Image Quality.