Klasifikasi Penyakit Daun Tomat Menggunakan Metode Convolutional Neural Network Berbasis Mobile di Pertanian Desa Bakalan

Dewi, Heni Lusiana (2025) Klasifikasi Penyakit Daun Tomat Menggunakan Metode Convolutional Neural Network Berbasis Mobile di Pertanian Desa Bakalan. Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (cover) 21082010060.-cover.pdf Download (2MB)
	Text (bab 1) 21082010060.-bab1.pdf Download (274kB)
	Text (bab 2) 21082010060.-bab2.pdf Restricted to Repository staff only until 30 June 2027. Download (912kB)
	Text (bab 3) 21082010060.-bab3.pdf Restricted to Repository staff only until 30 June 2027. Download (376kB)
	Text (bab 4) 21082010060.-bab4.pdf Restricted to Repository staff only until 30 June 2027. Download (1MB)
	Text (bab 5) 21082010060.-bab5.pdf Download (249kB)
	Text (daftar pustaka) 21082010060.-daftarpustaka.pdf Download (197kB)
	Text (lampiran) 21082010060.-lampiran.pdf Restricted to Repository staff only Download (336kB)

Abstract

Diseases affecting tomato leaves are one of the main factors that can reduce crop yields and harm farmers in Bakalan Village, Sidoarjo. Therefore, the application of artificial intelligence technology, especially in the field of digital image processing, is a potential solution for automatically and accurately identifying leaf diseases. This study aimed to compare the performance of five pretrained Convolutional Neural Network (CNN) architectures, namely DenseNet121, EfficientNetB0, InceptionV3, MobileNetV2, and ResNet50, in classifying tomato leaf diseases. The models applied transfer learning techniques to make the training process more efficient. Each model was trained using three different optimizers: Adam, RMSprop, and SGD, to determine the best combination of architecture and optimization technique for classifying tomato leaf image datasets. Model performance was evaluated based on validation data. The best-performing models were then tested on unseen test data that were not used during training. The results showed that the MobileNetV2 model trained with the RMSprop optimizer achieved the highest classification performance, with an accuracy of 0.9972, precision of 0.9973, recall of 0.9972, and an F1-score of 0.9972. These metrics indicated that the model was capable of performing classification optimally. The best-performing model was then implemented in a mobile application using the Flutter framework. This application classifies types of tomato leaf diseases based on input images and provides diagnosis results along with treatment recommendations. It is expected that this application will assist farmers in Bakalan Village in quickly and easily identifying and handling tomato leaf diseases, thereby minimizing crop losses.

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
UNSPECIFIED	Arifiyanti, Amalia Anjani	NIDN0712089201	amalia_anjani.fik@upnjatim.ac.id
UNSPECIFIED	Najaf, Abdul Rezha Efrat	NIDN0029099403	rezha.efrat.sifo@upnjatim.ac.id

Subjects:

T Technology > T Technology (General)

Divisions:

Faculty of Computer Science > Departemen of Information Systems

Depositing User:

Heni Lusiana Dewi

Date Deposited:

30 Jun 2025 08:02

Last Modified:

30 Jun 2025 08:02

URI:

https://repository.upnjatim.ac.id/id/eprint/39136