Desarrollo de aplicativo Web para agricultura de precisión en empresas florícolas utilizando datos de cultivos de flores de la empresa “Agroscan”

Abstract

This project describes the development and implementation of a web application for precision agriculture in flower companies, with data from rose crops provided by the Agroscan company. This application will have embedded a time series classification model generated in a previous study, which uses daily measurements of three variables taken from a crop plot and returns as an output a recommended agricultural activity to be applied in the plot from which the data was taken. The application will have two types of users: agricultural technician and administrator. The agricultural technician will be the one who enters the data of the measurements in the web application, while the administrator will carry out the user management. The software development methodology used is Scrum. The application will be developed in Python language version 3.7 with the Flask framework, version 2.1.1, which is designed for the development of web applications. The classification model was built using the Keras library version 2.3.1. As a result, a web application with a Deep Learning model is obtained, which could facilitate the implementation of precision agriculture in rose plantations.