Aplikasi Image Enhancement untuk Peningkatan Kualitas Citra Ultrasonografi Ginjal

Abstract

. Combination of four filters namely gaussian filter, median filter, wiener filter, average filter were tested using two contrast enhancement techniques, intensity adjustment and histogram equalization, were tested to improve the quality of kidney ultrasound images. The research was conducted using 40 images, consist of 9 normal images, 17 hydronerfosis images, and 14 kidney stones images. The measured image quality were PSNR ( Peak Signal to Noise Ratio ) and MSE ( Mean Square Error ) . The higher the PSNR and the lower the MSE, the better the image quality. Visual evaluation through questionnaires to clinicians has also been carried out to assess the visualization of the kidney and its abnormalities. The results of PSNR and MSE calculations showed that every image processings methods combinations produce different results in each image categories. However, whether in normal, hydronefrosis, or kindey stone categories, the combination of filters with image adjustment method gave the highest PSNR and lowest MSE. Meanwhile, the results of the visual evaluation from the clinicians showed that the best image enhancement technique in improving the visualization of abnormalities in kidney ultrasound images ( Hydronephrosis and kidney stones ) was the combination of a wiener filter with intensity adjustment, in accordance with the results of the PSNR measurement .