SANTOSA, MOCHAMMAD KEVIN (2023) IMPLEMENTASI ARSITEKTUR ALEXNET DAN RESNET34 PADA KLASIFIKASI CITRA PENYAKIT DAUN KENTANG MENGGUNAKAN TRANSFER LEARNING. Undergraduate thesis, UPN VETERAN JAWA TIMUR.
|
Text (Cover)
19081010170-cover.pdf Download (1MB) | Preview |
|
|
Text (Bab1)
19081010170-bab1.pdf Download (20kB) | Preview |
|
![[img]](http://repository.upnjatim.ac.id/style/images/fileicons/text.png) |
Text (Bab 2)
19081010170-bab2.pdf Restricted to Registered users only until 23 November 2025. Download (289kB) |
|
|
Text (Bab 3)
19081010170-bab3.pdf Restricted to Registered users only until 23 November 2025. Download (490kB) |
|
|
Text (Bab 4)
19081010170-bab4.pdf Restricted to Registered users only until 23 November 2025. Download (6MB) |
|
|
Text (Bab 5)
19081010170-bab5.pdf Download (10kB) | Preview |
|
|
Text (Daftar Pustaka)
19081010170-daftarputaka.pdf Download (79kB) | Preview |
Abstract
This study aims to apply the AlexNet and ResNet34 architectures in classifying disease images on potato leaves using transfer learning. The objectives of the study are to evaluate the ability of these two architectures in identifying diseases on potato leaves and to find model settings that can optimize their results. The results showed that by using a number of epochs of 16, batch size 14, and optimizer Adam, both models were able to achieve a good balance between classification accuracy and overfitting control. Both models were able to distinguish healthy potato plants from those infected with early blight and late blight with excellent accuracy. In addition, this study also evaluated the data sharing of the AlexNet and RESNet34 models. The results show that the AlexNet model achieves the best performance with 80%-20% data split, while the ResNet34 model achieves the best results with 70%-30% data split. This research has important relevance in the context of potato farming, where potato leaf diseases pose a serious threat to crop yields and agricultural productivity. Technological solutions such as image processing are key in identifying diseases efficiently and providing timely treatment to infected plants.
| Item Type: | Thesis (Undergraduate) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||||||
| Subjects: | T Technology > T Technology (General) | ||||||||||||
| Divisions: | Faculty of Computer Science > Departemen of Informatics | ||||||||||||
| Depositing User: | Mochammad Kevin Santosa | ||||||||||||
| Date Deposited: | 23 Nov 2023 08:34 | ||||||||||||
| Last Modified: | 23 Nov 2023 08:34 | ||||||||||||
| URI: | http://repository.upnjatim.ac.id/id/eprint/18809 |
Actions (login required)
 |
View Item |