Medical Technology is growing by the growth of the era. The detection of illness which is suffered by human can be detected earlier by doing observation of the symptom that emerge from the sufferer. The observation of the symptom can be done to the human organs which are probably changing because of illness, for example: the retina of eye. The changing of the structure can be seen is the blood vessel which becomes bigger or the disorder of the blood vessel of the retina of eye. In order to detect the illness, initial stage is to perform segmentation of the blood vessel. This study is using 2D-Gabor filters for segmenting the image. It is divided into 2 stages, namely preprocessing and segmentation. In the early stage of preprocessing consists of taking the green channel image, and improve the image contrast by Histogram Equalization. And the second stage is segmentation by 2D-Gabor filter method, thresholding the image, clean up the image of the noise, and the field of view. Then the results of the process compared with the groundtruth image to calculate the level of accuracy. The test performed on a database of Digital Retinal Images for Vessel Extraction (DRIVE) as many as 20 images. The accuracy of the results obtained from this test was 81.11%. The image of the result of segmentation is quite good, so the 2D-Gabor filter can be properly segmenting.

Informasi yang terkait dengan IMPLEMENTASI SEGMENTASI PEMBULUH DARAH RETINA PADA CITRA FUNDUS MATA BERBASIS HISTOGRAM EQUALIZATION DAN 2D GABOR FILTER

Riset implementasi Mahadata