KLASIFIKASI VARIETAS BERAS MENGGUNAKAN CONVOLUTIONAL NEURAL NETWORK DENGAN ARSITEKTUR MOBILENETV2

YUDISTIRA, MUHAMMAD IKHSAN and Miraswan, Kanda Januar and Rizqie, M. Qurhanul (2025) KLASIFIKASI VARIETAS BERAS MENGGUNAKAN CONVOLUTIONAL NEURAL NETWORK DENGAN ARSITEKTUR MOBILENETV2. Undergraduate thesis, Sriwijaya University.

	Image RAMA_55201_09021382126148_cover.pdf - Cover Image Available under License Creative Commons Public Domain Dedication. Download (200kB)
	Text RAMA_55201_09021382126148.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_55201_09021382126148_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (9MB) | Request a copy
	Text RAMA_55201_09021382126148_0009019002_0203128701_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (783kB)
	Text RAMA_55201_09021382126148_0009019002_0203128701_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (292kB) | Request a copy
	Text RAMA_55201_09021382126148_0009019002_0203128701_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (248kB) | Request a copy
	Text RAMA_55201_09021382126148_0009019002_0203128701_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (498kB) | Request a copy
	Text RAMA_55201_09021382126148_0009019002_0203128701_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (275kB) | Request a copy
	Text RAMA_55201_09021382126148_0009019002_0203128701_06_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (43kB) | Request a copy
	Text RAMA_55201_09021382126148_0009019002_0203128701_07_lamp.pdf Available under License Creative Commons Public Domain Dedication. Download (77kB)

Abstract

Rice consists of various varieties withdistinctmorphological characteristics, yet manual classification remains subjective and reliant on expert judgment. This study develops an automated rice variety classification system using the Convolutional Neural Network (CNN) method with the MobileNetV2 architecture. The dataset includes 75,000 images of five rice varieties, divided into 70% training, 5% validation, and 25% testing. The model was trained using transfer learning, data augmentation, and callbacks such as ModelCheckpoint and EarlyStopping. The training results showed a validation accuracy of up to 99.09% with stable validation loss, indicating no overfitting. The model was evaluated using metrics such as accuracy, precision, recall, and F1-score, and implemented in a Streamlit-based application to provide an interactive classification interface. The results confirm that MobileNetV2 is an effective and efficient architecture for rice variety classification.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Ilmu Komputer, Teknik Informatika
Subjects:	T Technology > T Technology (General) > T58.4 Managerial control systems Information technology. Information systems (General)
Divisions:	09-Faculty of Computer Science > 55201-Informatics (S1)
Depositing User:	Muhammad Ikhsan Yudistira
Date Deposited:	18 Jun 2025 07:59
Last Modified:	18 Jun 2025 07:59
URI:	http://repository.unsri.ac.id/id/eprint/175757

Actions (login required)

View Item