Blood vessel segmentation in the retina fundus image becomes substantial in the medicine, because it can be used to detect some diseases, such as diabetic retinopathy, hypertension, and cardiovascular. Doctor took about two hours to detect the blood vessels of the retina. Therefore, screening methods are needed to make it faster. The previous methods was able to segment the blood vessels that are sensitive to variations of the width of blood vessels, but there is over-segmentation in the area of pathology. Therefore, this study aims to develop a seg-mentation method of blood vessels in retinal fundus images which can reduce over-segmentation in the area of pathology using Gradient Based Adaptive Thresholding and Region Growing. The proposed method consists of three stages, namely the segmentation of the main blood vessels, the pa-thology area detection and the segmentation of thin blood vessels. The Main blood vessels segmentation was using high-pass filtering and top-hat reconstruction on the green channel which used an adjusted contras image that pro-duce a clear distinction between object and background. The pathology area detection was using Gradient Based Adaptive thresholding method. The Thin blood vessels segmentation was using Region Growing based on the infor-mations that was gained from the main blood vessel segmentation and the pathology area detection. The Output of the main blood vessel and thin blood vessels segmentation were then combined to reconstruct a segmented image of the blood vessels as the system output. This method is able to segment the retinal blood vessel in DRIVE fundus image by the accuracy of 95.25% and Area Under Curve (AUC) value in the relative operating characteristic curve (ROC) of 74.28%.

Original languageIndonesian
Pages (from-to)105-116
Number of pages12
JournalRegister: Jurnal Ilmiah Teknologi Sistem Informasi
Issue number2
Publication statusPublished - Jul 2016


  • Blood vessel
  • Fundus retina image
  • Gradient based adaptive thresholding
  • Pathology
  • Region growing
  • Segmentation

Cite this