Kartikasari, Rani Dwi (2023) IMPLEMENTASI ALGORITMA CONVOLUTIONAL NEURAL NETWORK (CNN) UNTUK IDENTIFIKASI JENIS TANAMAN RIMPANG (ZINGIBERACEAE). Skripsi (S1) thesis, Universitas Muhammadiyah Ponorogo.
![[img]](http://eprints.umpo.ac.id/style/images/fileicons/text.png) |
Text (SURAT PERSETUJUAN UNGGAH KARYA ILMIAH)
Surat Persetujuan Unggah Karya.pdf Download (223kB) |
|
Text (HALAMAN DEPAN)
HALAMAN DEPAN.pdf Download (2MB) |
|
Text (BAB 1)
BAB 1.pdf Download (186kB) |
|
Text (BAB 2)
BAB 2.pdf Restricted to Repository staff only Download (467kB) |
|
Text (BAB 3)
BAB 3.pdf Restricted to Repository staff only Download (553kB) |
|
Text (BAB 4)
BAB 4.pdf Restricted to Repository staff only Download (453kB) |
|
Text (BAB 5)
BAB 5.pdf Restricted to Repository staff only Download (172kB) |
|
Text (DAFTAR PUSTAKA)
DAFTAR PUSTAKA.pdf Download (227kB) |
|
Text (LAMPIRAN)
LAMPIRAN.pdf Restricted to Repository staff only Download (3MB) |
|
Text (SKRIPSI FULL TEXT)
SKRIPSI RANI DWI KARTIKASARI.pdf Restricted to Repository staff only Download (6MB) |
Abstract
Rhizomes (Zingiberaceae) are plants with modified stems that creep below the surface of the soil and are capable of producing new shoots and roots from their nodes. Rhizome plants (Zingiberaceae) are known as various types of "temu" or "empon-empon," including Ginger, Kaempferia galanga, White Turmeric, Turmeric, Galangal, Greater Galangal, Curcuma xanthorrhiza, Black Turmeric, Fingerroot, and Mango Ginger. The numerous types of rhizome plants (Zingiberaceae) with similar characteristics often confuse the general public, especially teenagers, making it difficult to differentiate between them. In this study, a Convolutional Neural Network (CNN) algorithm with the VGG19 architecture was used to facilitate the identification of rhizome plant species (Zingiberaceae) for the public. This research utilized a dataset of 1000 rhizome images, divided into training, testing, and validation data. The rhizome images underwent data preprocessing by resizing them from 500 x 500 to 200 x 200 pixels. In the model design phase, three scenarios were tested: Scenario 1 with a dataset ratio of 85:10:5, 25 epochs, and a batch size of 30; Scenario 2 with a dataset ratio of 80:10:10, 20 epochs, and a batch size of 32; and Scenario 3 with a dataset ratio of 70:20:10, 10 epochs, and a batch size of 20. The results showed that Scenario 2 performed the best with an accuracy of 90%, a loss of 0.285, precision of 93%, recall of 89%, and an F1-Score of 91%. However, despite achieving 90% accuracy, the model's image predictions reached only 56% out of 100 testing data.
| Item Type: | Thesis (Skripsi (S1)) |
|---|---|
| Uncontrolled Keywords: | Convolutional Neural Network (CNN), Identifikasi, Jenis Tanaman Rimpang, VGG19 |
| Subjects: | T Technology > TK Electrical engineering. Electronics Nuclear engineering |
| Divisions: | Faculty of Engineering > Department of Informatic Engineering |
| Depositing User: | ft . userft |
| Date Deposited: | 12 Sep 2023 04:59 |
| Last Modified: | 12 Sep 2023 04:59 |
| URI: | http://eprints.umpo.ac.id/id/eprint/12595 |
Actions (login required)
 |
View Item |
