| Autor: |
Gaikwad, Varsha P., Musande, Vijaya |
| Předmět: |
|
| Zdroj: |
AIP Conference Proceedings; 2023, Vol. 2822 Issue 1, p1-11, 11p |
| Abstrakt: |
Since decade agriculture is the main source of food source. It plays a very vital role in food production. More than sixty percent of the population of earth depends on agriculture sources for their main supply. Various factors such as climate, rain, soil, and insects affects on growth of the plant. Every year more than 15-20 percent crop losses occurs due to plant disease. Potato and tomato mostly used in every kitchen as well as in various industrial productions. These crops are mostly affected with late and early blight disease. The late and inaccurate identification of disease decreases the productivity of the crops. The key purpose of our research is to discover plant disease at an early stage by using the Convolution Neural Network (CNN). For this research we considered Plantvillage dataset of 3580 tomato plant leaf images and 2150 potato plant leaf images. We used CNN customized models. For our research AlexNet model achieves 84.7% accuracy and VGG16 model achieves 96.5% accuracy for tomato and potato plant leaves respectively. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
