Please use this identifier to cite or link to this item: https://repositori.uma.ac.id/handle/123456789/24588
Full metadata record
DC FieldValueLanguage
dc.contributor.advisorKhairina, Nurul-
dc.contributor.authorPurba, Sentia Ovania-
dc.date.accessioned2024-07-11T03:59:48Z-
dc.date.available2024-07-11T03:59:48Z-
dc.date.issued2024-03-27-
dc.identifier.urihttps://repositori.uma.ac.id/handle/123456789/24588-
dc.description95 Halamanen_US
dc.description.abstractMata merupakan alat indra makhluk hidup yang sangat penting khususnya pada manusia. Dimana jika rentina mata mengalami kerusakan atau terkena penyakit, khususnya penyakit mata katarak maka kualitas rentina mata tidak dapat digunakan dengan baik. Maka diperlukan pendekatan digital agar dapat mengenali mata tersebut positif katarak atau negatif secara cepat, tepat, efesien dan akurat. Sehingga penelitian ini membuat suatu investigasi penyakit mata manusia khususnya katarak melalui citra rentina mata kedalam bentuk klasifikasi dengan cepat, efesien, dan akurat. Klasifikasi yang dilakukan dalam penelitian ini menggunakan model algoritma Vision Transformer dari CNN. Berdasarkan hasil training pada enam skenario yang diuji. dengan menggunakan dataset training memanfaatkan Hyperparameter Epoch 25 dan 50, Input Shape 224x224x3 (RGB channel), batch size 32 dan Optimizer (Adam, RMSprop, SGD) dan Learning Rate 0.001 dengan dataset 1120 citra rentia. Hasil dari penelitian ini dengan menggunakan Vision Transformer yang memanfaatkan Hyperparameter Epoch 25 dan 50, Input Shape 224x224x3 (RGB channel), batch size 32 dan Optimizer (Adam, RMSprop, SGD) dan Learning Rate 0.001 mendapatkan hasil secara berturut-turut, dimana akurasi terbaik yaitu Akurasi 92.86%%, Precision 100%, Recall 85.72%, F1-score 92,31%. The eyes are a very important sense organ in living creatures, especially humans. Where if the eye retina is damaged or affected by disease, especially cataracts, the quality of the eye retina cannot be used properly. So a digital approach is needed to be able to identify whether the eye is positive for cataracts or negative quickly, precisely, efficiently and accurately. So this research makes an investigation of human eye diseases, especially cataracts, through images of the retina of the eye in a fast, efficient and accurate classification form. The classification carried out in this research uses the Vision Transformer algorithm model from CNN. Based on training results on the six scenarios tested. using a training dataset utilizing Hyperparameter Epoch 25 and 50, Input Shape 224x224x3 (RGB channel), batch size 32 and Optimizer (Adam, RMSprop, SGD) and Learning Rate 0.001 with a dataset of 1120 rentia images. The results of this research using Vision Transformer which utilizes Hyperparameter Epoch 25 and 50, Input Shape 224x224x3 (RGB channel), batch size 32 and Optimizer (Adam, RMSprop, SGD) and Learning Rate 0.001 get results respectively, where the best accuracy namely Accuracy 92.86%%, Precision 100%, Recall 85.72%, F1-score 92.31%.en_US
dc.language.isoiden_US
dc.publisherUniversitas Medan Areaen_US
dc.relation.ispartofseriesNPM;198160066-
dc.subjectKlasifikasien_US
dc.subjectPenyakit Mataen_US
dc.subjectKataraken_US
dc.subjectConvolutional Neural Networken_US
dc.subjectPengolah Dasar Transformeren_US
dc.subjectVision Transformeren_US
dc.titleKlasifikasi Penyakit Mata pada Manusia dengan Menggunakan Model Arsitektur Vision Transformersen_US
dc.title.alternativeClassification of Eye Diseases in Humans Using the Vision Transformers Architectural Modelen_US
dc.typeThesisen_US
Appears in Collections:SP - Informatic Engineering

Files in This Item:
File Description SizeFormat 
198160066 - Sentia Ovania Purba - Chapter IV.pdf
  Restricted Access
Chapter IV1.7 MBAdobe PDFView/Open Request a copy
198160066 - Sentia Ovania Purba - Fulltext.pdfCover, Abstract, Chapter I, II, III, V, Bibliography1.89 MBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.