Please use this identifier to cite or link to this item: http://hdl.handle.net/10609/132028
Title: Sistema de monitorización y detección de plagas en cultivos aplicando algoritmos de Deep Learning
Author: Ferrer Martínez, Claudia
Director: Monzo, Carlos  
Tutor: Crespo García, David
Abstract: In this project, a global design of a field monitoring system is presented, in which, using sensors and different types of data collection devices, UAVs and cameras for obtaining images, environmental data of the land and plants is constantly monitored, with the aim of applying precision agriculture techniques and being able to optimize processes such as irrigation, fertilization, the use of pesticides or harvesting, in addition to detecting possible diseases and pests in the plants. First, a design of the architecture of the smart agriculture system is proposed, then, different types of data acquisition devices are theoretically analyzed as well as the protocols and communication systems required for data transmission at each stage, and a specific solution is proposed for each of the monitoring systems. Secondly, a system for detecting pests or diseases in the crop is implemented, applying deep learning techniques, specifically, a convolutional neural network trained from scratch is implemented with a dataset of images of leaves from different kind of crops, which classifies the images into the corresponding crop type and pest type, and their performance is compared with pretrained neural networks.
Keywords: internet of things
deep learning
precision agriculture
sensors
pest detection
convolutional neural networks
Document type: info:eu-repo/semantics/masterThesis
Issue Date: May-2021
Publication license: http://creativecommons.org/licenses/by-nc-nd/3.0/es/  
Appears in Collections:Bachelor thesis, research projects, etc.

Files in This Item:
File Description SizeFormat 
cferrermartiTFM0621memoria.pdfMemoria del TFM25,16 MBAdobe PDFThumbnail
View/Open
cferrermartiTFM0621presentación.pdfPresentación del TFM2,97 MBAdobe PDFThumbnail
View/Open