Deteksi Penyakit Pada Daun Kentang Menggunakan Pengolahan Citra dengan Metode Convolutional Neural Network

Abstract

Produk pertanian kentang menjadi sangat penting karena termasuk makanan utama bagi manusia. Kentang memiliki kandungan karbohidrat yang menjadikanya sebagai makanan utama. Dalam mengelola pertanian kentang ini tentu memiliki beberapa kendala diantaranya adalah penyakit yang menyerang pada daun kentang yang jika dibiarkan akan menghasilkan produksi yang buruk atau bahkan gagal panen. Late blight dan early blight adalah penyakit yang sering ditemui pada daun kentang. Penyakit ini memiliki gejala masing-masing sehingga para petani dapat melakukan pencegahan jika melihat gejala pada daun kentang, tetapi langkah ini memliki kelemahan yaitu proses identifikasi yang lama, dan jika penanganan pada penyakit daun ini sangat lambat akan mengakibatkan penambahan biaya perawatan. Dengan memanfaatkan teknologi yaitu berupa pengolahan citra digital maka hal ini bisa diatasi, jadi pada penelitian ini akan mengusulkan metode yang tepat dalam mendeteksi penyakit pada daun kentang. Klasifikasi akan dilakukan dengan tiga kelas berupa daun sehat, early blight, dan late blight menggunakan metode Deep Learning mengguanakan arsitektur Convolutional Neural Network (CNN). Hasil pada peneltian ini dianggap baik karena pada epoch ke 10 dengan batch size 20 menghasilkan training akurasi 95% dan validation accuracy 94%.Kata Kunci—Penyakit daun kentang, late blight, early blight, identifikasi, CNNPotato agricultural products are essential because they are the leading food. Potatoes have carbohydrate content, which makes them the leading food for humans. But in carrying out this potato farming certainly has several obstacles, including the disease that attacks the potato leaves which if left unchecked will result in poor production or even crop failure. late blight and early blight are diseases that are often found in potato leaves. This disease has its own symptoms so that farmers can take precautions if they see symptoms on potato leaves, but this step has a weakness that is a long identification process, and if the handling of this leaf disease is very slow will result in additional maintenance costs. By utilizing technology in the form of digital image processing, this can be overcome, so this research will propose an appropriate method in detecting diseases in the leaves of potato plants. Classification will be carried out with three classes in the form of healthy leaves, early blight, and late blight using the Convolutional Neural Network (CNN) algorithm. The results of this research are considered good because on the 10th epoch with batch size 20 produces 95% accuracy training and 94% validation accuracy.Keywords—Potato leaf disease, late blight, early blight, identification, CNN

Rice Plant Disease Detection with Data Augmentation Using Transfer Learning

“…Epoch is a process of training in NN (Neural Network) in one round when the entire dataset goes back to the initial stage of this process. If only using 1 epoch in training data with a neural network model, it will be too large and burden the training process because the data used is quite a lot, so data rate per batch division or commonly called batch size is needed (Rozaqi, Sunyoto, &Arief, 2021).…”

“…In addressing this issue, information technology can be utilized to identify diseases by using digital image processing. The use of image processing in plant disease detection can assist agricultural managers in taking effective and efficient steps (Rozaqi, Sunyoto, &Arief, 2021).…”

Identifikasi Penyakit pada Citra Daun Kentang Menggunakan Convolutional Neural Network (CNN)

“…Penyakit yang paling umum ditemukan pada daun tanaman kentang yaitu busuk daun (late blight) dan bercak kering (early blight). Salah satu faktor penyebab terjadinya penyakit busuk daun berkembang yaitu tempat dingin dan lembab [4]. Dengan melakukan identifikasi kedua penyakit pada daun tanaman kentang ini diharapkan para petani dapat mencegah terjadinya serangan penyakit pada tanaman kentang, sehingga tidak akan terjadi lagi keterhambatan produksi dan pengurangan hasil produksi.…”

Leaf Image Identification: CNN with EfficientNet-B0 and ResNet-50 Used to Classified Corn Disease

“…Previous research related to the classification of corn leaf images has been carried out by Abdul Jalil Rozaqi, Andi Sunyoto, and Rudyanto Arief (2021) with the title "Detection of Diseases in Potato Leaves Using Image Processing with the Convolutional Neural Network Method" proposing the application of the algorithm used is the Convolutional Neural Network, with the number of datasets is 1152 and divided into 4 classification classes, using a dataset size of 150x150 with a sharing ratio of 80% training data and 20% validation data to get 94% accuracy value [20].…”