Penerapan Metode Convolutional Neural Network (CNN) Dalam Mengklasifikasikan Penyakit Daun Tanaman Padi

Christiawan, Gracia Yoel
Putra, Roy Andani
Sulaiman, Azis
Poerbaningtyas, Evy
Putri Listio, Syntia Widyayuningtias

J-Intech (Journal of Information and Technology)
Vol 11 No 2 (2023): J-Intech : Journal of Information and Technology

Rice is a staple crop in Indonesia. Most farmers choose rice as the main crop for agricultural land. Starting from the land to the tropical climate that occurs in Indonesia, it is very suitable for rice plants. Among these supports arise obstacles faced by farmers. Rice leaf diseases include Brownspot, Blas, Bacterial Leaf Blight (HDB). Classification of these diseases can be done using the CNN (Convolutional Neural Network) method. So far, the detection process for rice plant leaf diseases has been done manually. The CNN method can detect images from pixel to pixel so it is considered effective for detecting disease from images alone. This research uses a dataset of 1630 data which is divided into 3 disease classes. This research compares the number of epochs and uses the CNN InceptionV3 architecture. The results of this research show very good results with a lift of 98% with data that is not overfitting.