MAULANA IQBAL, RISYAD (2025) ANALISIS KLASIFIKASI VARIETAS BERAS MENGGUNAKAN METODE CONVOLUTIONAL NEURAL NETWORK (CNN). S1 thesis, Fakultas Teknik Universitas Sultan Ageng Tirtayasa.
![[img]](https://eprints.untirta.ac.id/style/images/fileicons/text.png) |
Text (Fulltext)
Risyad Maulana Iqbal _3332180024_Full Text.pdf Restricted to Registered users only Download (4MB) | Request a copy |
|
Text (Bab 1)
Risyad Maulana Iqbal_3332180024_01.pdf Restricted to Registered users only Download (1MB) | Request a copy |
|
Text (Bab 2)
Risyad Maulana Iqbal_3332180024_02.pdf Restricted to Registered users only Download (601kB) | Request a copy |
|
Text (Bab 3)
Risyad Maulana Iqbal_3332180024_03.pdf Restricted to Registered users only Download (94kB) | Request a copy |
|
Text (Bab 4)
Risyad Maulana Iqbal_3332180024_04.pdf Restricted to Registered users only Download (1MB) | Request a copy |
|
Text (Bab 5)
Risyad Maulana Iqbal_3332180024_05.pdf Restricted to Registered users only Download (27kB) | Request a copy |
|
Text (Daftar Referensi)
Risyad Maulana Iqbal_3332180024_Ref.pdf Restricted to Registered users only Download (170kB) | Request a copy |
|
Text (Lampiran Skripsi)
Risyad Maulana Iqbal_3332180024_Lamp.pdf Restricted to Registered users only Download (1MB) | Request a copy |
|
Text (Dokumen Hasil Cek Plagiasi)
Risyad Maulana Iqbal_3332180024_CP.pdf Restricted to Registered users only Download (3MB) | Request a copy |
Abstract
Rice Variety Analysis Using the Convolutional Neural Network Method (CNN) Rice is a staple food that is widely consumed by the community. The high level of need is widely used by business actors to commit fraud in the form of mixing. This can harm the community economically and contrary to existing consumer protection regulations. Convolutional Neural Network is a method used for image data identifying image objects. Input rice image measuring 250x250 pixels from 7 varieties with 75:25 split data division. Using google colabs CNN architecture from keras frameworks used are Vgg16Net, ResNet50, InceptionV3, and InceptionResNetV2. The results obtained are InceptionV3 architecture has the best results with an accuracy of 0.985 out of 345 correct predictions, InceptionResNetV2 with an accuracy of 0.982 out of 344 correct predictions, Vgg16Net with an accuracy of 0.98 out of 343 correct predictions, and ResNet50 with an accuracy of 0.957 out of 335 correct predictions. Convolutional Neural Network architecture gets good performance in the process of classifying varieties, with the help of transfer learning stages the use of learning models can be done efficiently and quickly. Word Keys: Rice, Deep Learning, Convolutional Neural Network, Transfer Learning
| Item Type: | Thesis (S1) | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
|||||||||
| Additional Information: | Analaisis Klasifikasi Varietas Beras Menggunkan Metode Convolutional Neural Network (CNN) Beras merupakan bahan pangan pokok yang banyak dikonsumsi masyarakat. Tingkat kebutuhan yang tinggi banyak dimanfaatkan pelaku usaha untuk melakukan kecurangan berupa pencampuran. Hal tersebut dapat merugikan masyarakat secara ekonomi dan berlawanan dengan peraturan perlindungan konsumen yang ada. Convolutional Neural Network merupakan metode yang digunakan untuk data image pengidentifikasian objek gambar. Input citra beras berukuran 250x250 pixel dari 7 varietas dengan pembagian split data 75:25. Menggunakan google colabs arsitektur CNN dari keras framework yang digunakan ialah Vgg16Net, ResNet50, InceptionV3, dan InceptionResNetV2. Hasil yang didapatkan yaitu arsitektur InceptionV3 memiliki hasil terbaik dengan akurasi 0.985 dari 345 prediksi benar, InceptionResNetV2 dengan akurasi 0.982 dari 344 prediksi benar, Vgg16Net dengan akurasi 0.98 dari 343 prediksi benar, dan ResNet50 dengan akurasi 0.957 dari 335 prediksi benar. Arsitektur Convolutional Neural Network mendapatkan performa baik dalam proses pengklasifikasian varietas, dengan dibantu tahapan transfer learning penggunaan model learning dapat dilakukan secara efisien dan cepat. Kata Kunci: Beras, Pembelajaran Mendalam, Jaringan Syaraf Konvolusi, Pembelajaran Transfer. | |||||||||
| Uncontrolled Keywords: | Word Keys: Rice, Deep Learning, Convolutional Neural Network, Transfer Learning | |||||||||
| Subjects: | T Technology > TK Electrical engineering. Electronics Nuclear engineering | |||||||||
| Divisions: | 03-Fakultas Teknik > 20201-Jurusan Teknik Elektro | |||||||||
| Depositing User: | Risyad Maulana Iqbal | |||||||||
| Date Deposited: | 11 Jul 2025 03:21 | |||||||||
| Last Modified: | 11 Jul 2025 03:21 | |||||||||
| URI: | http://eprints.untirta.ac.id/id/eprint/51201 |
Actions (login required)
 |
View Item |