| Autor: |
Adi, Kusworo, Widodo, Catur Edi, Widodo, Aris Puji, Setiadi, Agus, Setiawan, Aji |
| Předmět: |
|
| Zdroj: |
Mathematical Modelling of Engineering Problems; May2024, Vol. 11 Issue 5, p1279-1285, 7p |
| Abstrakt: |
Rice plants (Oryzae Sativa sp.) are rice-producing food plants which play an important role in economic life in Asian countries, especially Indonesia. This primary need is very difficult to replace with other staples, such as corn, tubers, sago and other sources of carbohydrates. However, global climate change which has an impact on climate anomalies causes rice diseases to develop rapidly. The aim of this research is to create an application on an Android-based cell phone that can identify and classify rice plant diseases using the Convolutional Neural Network (CNN) method. The research material was 1600 images of rice leaves consisting of 4 classes, namely 400 images of rice affected by Leaf Blight disease, 400 images of rice affected by Brown Spot disease, 400 images of rice affected by Leaf Smut disease, and 400 images of healthy rice. In each class, the images are divided into 380 for training and 20 for testing. During the rice image training process using the Python programming language, the accuracy results were 83%. Then the results are saved in the form of a model file, and entered as training data into the program in Android Studio to be used as an application. Testing the application with 20 rice leaf images for each class resulted in an actual accuracy of 94%. This application can also be run on all versions of Android. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
