Text

IMPLEMENTASI AUGMENTASI DAN MODIFIKASI ARSITEKTUR DENSENET-DWS DALAM KLASIFIKASI PENYAKIT GLAUKOMA PADA CITRA RETINA

---

Penilaian

---

Convolutional Neural Network (CNN) is a method that is often used in the process of image classification. CNN requires large and balanced data for each class in the training process, but the data used in this study is still relatively small and the number in each class is still not balanced, so we need a method to increase data, namely augmentation. DenseNet architecture is part of the CNN method that can be used to classify glaucoma disease, but DenseNet applies dense network exploitation that can increase the number of parameters, thereby reducing computational efficiency. Addition of depthwise separable convolution and dense layer at the end of the model will again increase the computational efficiency. The stages of the research carried out include data description, data augmentation, image repaired, training, testing, model evaluation, analysis and interpretation of results, and conclusion. The results of the evaluation of the model obtained are the values of accuracy, sensitivity, specificity, f1-score, and Cohens kappa respectively 90.32%, 85.63%, 92.82%, 85.45%, and 78.23%. Based on the results of the evaluation of the model, it was found that the augmentation and modification of the DenseNet architecture was able to classify the glaucoma disease in the image given well.

---

Availability

Inventory Code	Barcode	Call Number	Location	Status
2207002965	T76762	T767622022	Central Library (Referens)	Available but not for loan - Not for Loan

Detail Information

Series Title

-

Call Number

T767622022

Publisher

Inderalaya : Jurusan Matematika, Fakultas Matematika Dan Ilmu Pengetahuan Alam, Universitas Sriwijaya., 2022

Collation

xii, 59 hlm.; ilus.; 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

518.107

Content Type

-

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Algoritma
Jurusan Matematika

Specific Detail Info

-

Statement of Responsibility

SEPTA

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment