Deep learning como ferramenta na detecção e classificação de doenças agrícolas em folhas

Abstract

The com culture is present in the lives of Brazilians in different ways. Itú usa Rangers from human food (boiled corn, cake, mush, etc.) to the use of grais to feed pias and poultry. A factor of concern, not only corn producers, but farmers in general, are agricultural diseases that cause large losses in the harvest, decreasing profit or even causing losses. This project aimed to develop a system capable of recognizing diseases that are present in the plantation. For that, artificial intelligence tecniques were used throught Convolutional Neural Networks. These networks were trained, tested and evaluated in order to reach a conclusion about their reliability.