Klasifikasi Penyakit Tanaman Jagung Menggunakan Metode Convolutional Neural Network (CNN)

Abstract

Corn is one of the cereal crops and an important food ingredient because it contains a source of carbohydrates. It is the third largest food crop in the world after rice and wheat. In Indonesia, corn is one of the second largest food crops after rice. This plant has high yields and wide uses. However, in terms of disease, there are still many farmers who still have difficulty in distinguishing diseases, therefore digital assistance is needed to distinguish the characteristics of these plant diseases. Diseases used in this study were: blight and leaf rust. The dataset used has 2 types of corn plant diseases and a total of 2000 images of corn diseases. Convolutional Neural Network (CNN) algorithm is used for classification of corn plant diseases, which is included in the Deep Learning method which has good ability to recognize and classify a digital image object. This study uses the python programming language and the Tensorflow framework to conduct training and data testing. In this study, the classification of corn plant diseases using the Convolutional Neural Network (CNN) method was obtained with a total training accuracy of 97.5%, on data validation it got 100% accuracy and the level of accuracy in testing data using new data was 94%.