Detecção de doenças em folhas de soja utilizando redes neurais convolucionais

Abstract

The disease detection is vital to increase the productivity and quality of soybean cultivation. the detection of the presence of diseases it is usually carried out in a laboratory, which is time-consuming and costly. To overcome these issues, there is a growing demand for technologies that aim at a faster detection and classification of diseases. In this context, this work is the use of convolutional neural networks in combination with support vector machines to analyze and classify soybean leaf textures. In this approach, a set of data containing samples with evidence of diseases commonly observed in soybean crops was analized - Mildew, Oidio , Mosaic Virus, Rust and Bacterial Blight , achieving precisions greater than 90%.