Muhamad Ghozi Wijaya
Universitas Muhammadiyah Prof. DR. HAMKA

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

IDENTIFIKASI PENYAKIT TANAMAN PANGAN MENGGUNAKAN ALGORITMA CNN DENGAN METODE TRANSFER LEARNING MOBILENETV2 BERBASIS ANDROID Muhamad Ghozi Wijaya
Jurnal Teknik Informatika dan Komputer Vol. 4 No. 1 (2025): Jurnal Teknik Informatika dan Komputer
Publisher : Universitas Muhammadiyah Prof. DR. HAMKA

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22236/jutikom.v4i1.24789

Abstract

Food production in Indonesia faces various challenges, including pests and plant diseases, which can threaten sustainability and food security. This research aims to develop a disease identification model for rice, corn, and potato plants using convolutional neural networks (CNN) with MobileNetV2 transfer learning. The dataset used was obtained from Kaggle and processed through data augmentation to enhance image variation. MobileNetV2 was chosen for its efficiency and good performance on resource-constrained devices. The model training process utilized MobileNetV2 as the backbone, with the last 30 layers being trainable while the rest were frozen. The model was compiled using the adam optimizer and categorical crossentropy as the loss function, and it was trained for 13 epochs with a batch size of 32 using data augmentation. L2 regularization and a 0.5 dropout rate were applied to prevent overfitting, and early stopping was used to halt training if the validation loss did not improve after 5 epochs. The model was evaluated using a confusion matrix and classification report, achieving a validation accuracy of 91.03%. Subsequently, the model was integrated into an Android-based application designed to help farmers detect plant diseases. The application was assessed in terms of interface usability and willingness to use it in the future using the system usability scale (SUS), with testing results showing an average score of 70.1, placing it in category B (Excellent).