| Autor: |
Fedorchenko, Ievgen, Yusof, Mohd Faizal, Oliinyk, Andrii, Chornobuk, Maksym, Khokhlov, Mykola, Jamil Alsayaydeh, Jamil Abedalrahim |
| Předmět: |
|
| Zdroj: |
International Journal of Electrical & Computer Engineering (2088-8708); 2024, Vol. 14 Issue 5, p5866-5873, 8p |
| Abstrakt: |
The paper addresses the issue of automating the detection of fungal diseases in plants using digital images of their leaves. The spread of diseases among agricultural and horticultural crops causes significant economic losses worldwide, making the development of an effective and affordable solution to this problem highly valuable. Literature analysis suggests the viability of employing a convolutional neural network (CNN) to tackle this issue. The 'Fungus recognition' model was developed based on a custom CNN architecture using the TensorFlow library. The model underwent training and testing on a publicly available dataset. Test results show that 'Fungus recognition' achieves a classification accuracy level of 90%, surpassing similar models considered. The developed model can be adapted for deployment on mobile computing devices, paving the way for its practical implementation in agriculture and horticulture. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
