IMPLEMENTASI DEEP LEARNING UNTUK MENDETEKSI PENYAKIT PADA CITRA DAUN TANAMAN PADI JENIS CIHERANG, IR64, DAN INPARI 32 DI KABUPATEN MADIUN MENGGUNAKAN METODE CNN

Hidayat, Irfan Fajar (2026) IMPLEMENTASI DEEP LEARNING UNTUK MENDETEKSI PENYAKIT PADA CITRA DAUN TANAMAN PADI JENIS CIHERANG, IR64, DAN INPARI 32 DI KABUPATEN MADIUN MENGGUNAKAN METODE CNN. S1 thesis, Universitas PGRI Madiun.

	Text HALAMAN DEPAN.pdf Download (575kB)
	Text ABSTRACT.pdf Download (168kB)
	Text BAB 1.pdf Download (211kB)
	Text BAB 2.pdf Download (323kB)
	Text BAB 3.pdf Restricted to Repository staff only Download (308kB) | Request a copy
	Text BAB 4.pdf Restricted to Repository staff only Download (1MB) | Request a copy
	Text BAB 5.pdf Download (197kB)
	Text DAFTAR PUSTAKA.pdf Download (191kB)
	Text LAMPIRAN.pdf Restricted to Repository staff only Download (304kB) | Request a copy

Abstract

Irfan Fajar Hidayat, 2026. Implementation of Deep Learning for Detecting Diseases in Rice Leaf Images of Ciherang, IR64, and Inpari 32 Varieties in Madiun Regency Using CNN Method. Thesis. Informatics Engineering Study Program, Faculty of Engineering, Universitas PGRI Madiun. Advisor (I) Saifulloh, S.Kom., M.Kom. Co-Advisor (II) Pratiwi Susanti, S.Kom., M.MT. Rice plant is a strategic food commodity and the main source of carbohydrates for Indonesian society. The high rate of rice leaf disease attacks such as bacterial leaf blight, blast, brown spot, and tungro can cause harvest losses of up to 30–50% if not promptly identified and handled appropriately. The difficult process of conventional disease identification which relies on limited experts has caused a lack of ability among farmers to independently detect and handle rice diseases. With the rapid development of artificial intelligence technology, this study implements a deep learning model based on the MobileNetV2 architecture with a transfer learning approach to classify four classes of rice leaf diseases in Ciherang, IR64, and Inpari 32 varieties in Madiun Regency. The system is implemented as a Flutter-based Android application connected to a Flask inference server via REST API. The dataset used consists of 320 balanced images (80 images per class), comprising independently curated data across four classes (Bacterial Leaf Blight, Blast, and Tungro) with a 70:15:15 split ratio. Model testing results demonstrated an overall accuracy of 89.58% and a macro F1-score of 89.61% based on 48 test images. Usability testing using the System Usability Scale (SUS) method yielded an average score of 86.62, categorized as "Excellent" based on the scale established by Bangor et al. (2009)., indicating that the system is acceptable and usable by farmers and agricultural officers in the field.

Item Type:	Thesis/Skripsi/Tugas Akhir (S1)
Kata Kunci:	Deep Learning, MobileNetV2, Rice Leaf Disease, Transfer Learning, Flutter
Subjects:	T Technology > T Technology (General)
Divisions:	Fakultas Teknik > Teknik Informatika
Depositing User:	HIDAYAT FAJAR IRFAN
Date Deposited:	09 Sep 2026 07:57
Last Modified:	09 Sep 2026 07:57
URI:	http://eprint.unipma.ac.id/id/eprint/9613

Actions (login required)

View Item