Skripsi

PERBAIKAN KUALITAS DAN SEGMENTASI PEMBULUH DARAH CITRA RETINA DENGAN METODE CONTRAST STRETCHING DAN ADAPTIVE THRESHOLDING

---

Penilaian

---

One way to diagnose Diabetic retinopathy is to segment the blood vessels of the retinal image, however, retinal images obtained from the DRIVE and STARE datasets have varying contrasts, so image quality improvements are needed to obtain stable contrast results. The image quality improvement is carried out using the contrast stretching method and continued with the segmentation process using the adaptive thresholding method to obtain accurate segmentation results. The process of entering data is retinal fundus image, then green channel is extracted and carried out the contrast stretching process. The segmentation process is carried out using the adaptive thresholding method so as to produce a binary image which is then filtered. The output data in the form of retinal blood vessels. The datasets used are the DRIVE and STARE datasets. The results of the DRIVE dataset study resulted in an average value of 95.68% accuracy, 65.05% sensitivity, and 98.56% specificity. The results of the STARE dataset research using Adam Hoover's ground truth comparison resulted in an average accuracy value of 96.13% accuracy, 65.90% sensitivity, and 98.48% specificity, while the study using Valentina Kouznetsova's ground truth comparison resulted in an average accuracy value of 93.89% accuracy, 52.15% sensitivity, and 99.02% specificity. So it can be concluded that the results from the DRIVE and STARE datasets obtained good accuracy and specificity values, but this method still cannot detect fine blood vessels because the sensitivity values are still low.

---

Availability

Inventory Code	Barcode	Call Number	Location	Status
2107003452	T50640	T506402021	Central Library (REFERENCES)	Available but not for loan - Not for Loan

Detail Information

Series Title

-

Call Number

T506402021

Publisher

Inderalaya : Prodi Ilmu Matematika, Fakultas Matematika dan Ilmu Pengetahuan Alam., 2021

Collation

xv, 63 hlm. : ilus. ; 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

519.5307

Content Type

Text

Media Type

unmediated

Carrier Type

-

Edition

-

Subject(s)

Prodi Ilmu Matematika
Metode Contrast Stretching
Adaptive Thresholding

Specific Detail Info

-

Statement of Responsibility

DS

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment