Research Title: Efficient Enhancement of ultrasound images of abnormal kidney
Author: Haidar J. Ismaail and Azeez A. Barzinjy
Abstract: Recently, ultrasound imaging has become one of the most preferred imaging techniques amongst other than medical imaging techniques as it is reasonably priced, extensively accessible and relatively safe to user due to its safety radiation method. Nevertheless, the existence of speckle noise and low resolution in all ultrasound images affect the quality of the images which cause the difficulties in the interpretation of ultrasound images. Accordingly, ultrasound image enhancement technique is an essential process for superior disease diagnosis as well as for detection of pathological kidney damage precisely. In this study, five contrast enhancement techniques (namely; Contrast adjustment, Sharpening, Decorrelation stretch, Histogram Equalization, and Adaptive Histogram equalization) and three denoising filters (namely; Median, Wiener, and Bilateral) were optimized their parameters on images of variety kidney pathology (i.e. stone, benign, and cyst) to have maximum PSNR and minimum MSE values. The best individual techniques for contrast enhanced was sharpening whereas for denoising Median and Weiner filters were better than others. The results of combination of denoising (first step) and contrast enhancing (second step), and in reverse order, showed that Bilateral filter causes to decrease PSNR (so, it increases MSE) in a wide range (it decreases the quality of the images). The unpredicted results were found while adding Bilateral filtered images with Sharpening contrast enhancement images (hybrid technique) which show high values of PSNR (so, small MSE). So, for images required for both demonizing and contrasting that cause enhancement the Bilateral filter which add with sharpening will give promise results.
Download the PDF Document From Here
Proceeding Book of ICASEE
5th year , 7 – 9th April, 2018