Santosa, Atharizky Ade
Unknown Affiliation

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

Found 1 Documents
Search

Deteksi Penyakit pada Tanaman Padi Menggunakan Pengolahan Citra Digital dengan Metode Convolutional Neural Network Santosa, Atharizky Ade; Fu'adah, R Yunendah Nur; Rizal, Syamsul
JOURNAL OF ELECTRICAL AND SYSTEM CONTROL ENGINEERING Vol. 6 No. 2 (2023): Journal of Electrical and System Control Engineering
Publisher : Universitas Medan Area

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31289/jesce.v6i2.7930

Abstract

Rice plant is one of the important factors in supporting human life. When it starts to grow, of course, rice plants also often face problems such as pests or diseases that cause plants to die and lead to crop failure. So proper handling is needed to overcome the disease in rice plants. One of the treatments that can be done is by detecting diseases in rice plants, so that farmers can provide appropriate treatment for these problems. The research data will be processed through several stages, then the dataset used in this study consists of three classes of rice plant diseases, namely, bacterial leaf blight, brown spot, leaf smut and one class of healthy/healthy rice plants with a total of 16000 datasets collected from sources www.kaggle.net and previous research. The parameters tested in this study, namely hidden layer and optimizer affect system performance in the form of accuracy, precision, recall, fl-score, and loss values. In this study, the best results were obtained by using four hidden layers and Adam optimizer. Accuracy was 99.66%, precision, recall, fl-score was 99.66%. 100% and a loss of 0.0047 as well as a graph of the accuracy and loss performance in a good fit.