Machine Learning - Potato Leaf Disease Detection App (MR-PoLoD) - Android application for classifying healthy, early, and late blight

Aug 3, 2026

Upload by Liam

Potato production in Indonesia is rapidly increasing, yet farmers face substantial losses from plant diseases that spread when treatment is delayed. MR-PoLoD is an Android application designed to classify potato leaves into three categories: healthy, early blight, and late blight. Using a CNN-based machine learning model trained on 3165 images with a 70/15/15 split, the system achieves precision, recall, F1-score and accuracy values around 0.99 (99%), enabling faster disease identification to support timely farming actions.

LanguageEnglish

CategoryHealthcare

FormatPDF

Pages6

Reading Time15 min

AccessFree to download

Introduction

Potato production challenges and disease impact

Early blight and late blight symptoms and spread

Machine Learning-Potato Leaf Disease Detection App (MR-PoLoD) Ahmad Fauzi [1], Annisya E Chandra [2], Sofyah Imammah [3 Malvin Zapata [4], Marza I Marzuki [5], Soni Prayogi [6] * Department of Electrical Engineering, Faculty of Industrial Engineering, Universitas Pertamina [1], [2], [3], [4], [5], [6] Jakarta. Indonesia [soni.prayogi@universitaspertamina.ac.id](soni.prayogi@universitaspertamina.ac.id)* Abstract— Potato production in Indonesia has grown very rapidly, making Indonesia the largest potato producer in Southeast Asia. However, there are challenges for farmers in growing potatoes. Such as treating potatoes for various diseases. 2 diseases will occur in potato plants if not treated quickly, namely early blight disease caused by the fungus Alternaria solani and late blight disease caused by the microorganism Phytophthora infestans. The project "Potato Plant Leaf Disease Detector (MRPoLod)" aims to design an android application that can classify leaves on potato plants into 3 classifications, namely healthy, early, and late blight disease. This application uses the CNN (Convolutional Neural Network) Machine Learning Algorithm because currently, CNN is recognized as the most efficient and effective model in pattern and image recognition tasks. This application uses the Python programming language which is rich in library and framework availability so that it can meet the needs of machine learning and image classification tasks. The total data used for training data, data validation and data testing is 3165 images. With each division of the data process on the training data of 70%, validation of 15% & testing of 15% to test the effectiveness of the model that has been created. The performance of MRPoLod for each class, obtained a precision value, recall, and f1-score of 0.99. Likewise, the accuracy value achieved by the model is 0.99 or 99%. Thus, the expected application can facilitate farmers in classifying diseases on potato plant leaves. Keywords—Accuracy, application, CNN, healthy potatoes I. INTRODUCTION Indonesia is an agricultural country that has abundant water supplies, vast and fertile land, and produces various agricultural commodities. One of the agricultural commodities in Indonesia is potato farming. Potatoes are the fourth main food in the world, after rice, corn, and wheat. Potatoes are also one of the foods that contain carbohydrates. Potato production in Indonesia has grown very rapidly, making Indonesia the largest potato producer in Southeast Asia. However, there are challenges for farmers in growing potatoes. They face heavy losses every year due to various diseases, pests & extreme weather that attack the potato plants [1]. Recently in September 2023, 6 hectares of potato plants failed to harvest in the Dieng Valley, Central Java. The Head of Dieng Kulon Village, Batur District, Banjarnega Regency, Slamet Budiono said that the failed harvest was caused by the frost phenomenon due to extreme temperatures that occurred there, the temperature in June-September 2023 ranged from 13 ° -21 ° during the day and 3 ° - 12 ° at night and had frozen 5 times in the morning with temperatures of -1 ° to -3 ° which caused the frost to continue to increase [2] . The average age of their plants is currently between 40 days and-70 days, but the potato plants have not yet borne fruit and most of them are dry which causes material losses of up to hundreds of millions of rupiah [3] . Then, farmers reflected on the previous incident, they continued to replant potatoes on the slopes and covered the potato plants with grass, but this did not rule out the possibility that potatoes planted on the slopes could survive until harvest time due to extreme temperatures [4] . Farmers also said they could not quickly tell whether the potato plant leaves would die due to being attacked by the fungus Alternaria solani (Early Blight) which occurs in winter and can infect other leaves through rainwater, dew and direct c

FAQs

What diseases does MR-PoLoD detect on potato leaves?

MR-PoLoD targets early blight caused by Alternaria solani and late blight caused by Phytophthora infestans, in addition to a healthy category.

How does the MR-PoLoD application classify leaves?

The application uses a Convolutional Neural Network (CNN) machine learning approach to perform image-based classification of potato leaves.

What dataset and training split does the model use?

The project uses 3165 images total, divided into training (70%), validation (15%), and testing (15%) to evaluate model performance.

Aug 3, 2026

Upload by Liam