Deep Learning for Image-Recognition-Based Cassava Disease Detection

We trained a convolutional neural network on a cassava image database to identify three diseases and two types of pest damage: brown leaf spot, red mite damage, green mite damage, brown streak disease, and cassava mosaic disease. Results show that image recognition is an easily deployable strategy f...

Full description

Bibliographic Details
Main Author: CGIAR Platform for Big Data in Agriculture
Format: Informe técnico
Language:Inglés
Published: 2019
Subjects:
Online Access:https://hdl.handle.net/10568/122956
Description
Summary:We trained a convolutional neural network on a cassava image database to identify three diseases and two types of pest damage: brown leaf spot, red mite damage, green mite damage, brown streak disease, and cassava mosaic disease. Results show that image recognition is an easily deployable strategy for disease detection.