MESOI

Research Lines

Financing entity
Agencia de Gestió d'Ajuts Universitaris i de Recerca
Coordinating entity
La Salle
Period
Thursday, 23 July, 2020 to Friday, 23 April, 2021
Call
IdC 2019 - LLAVOR
Scope
Head of project
Raquel Ros
Research Group
Research Line
Project type

Agriculture is the World’s largest industry demanding 13M ha of fresh land every year. At present,  about 36% of the available land worldwide is exploited. Many claims that traditional farming system  do not meet the future demand for food, leaving us wondering how future farms will look like. 

 

Hydroponic farming is a method invented in the 1627 aimed at growing plants without soil. It offers great advantages over traditional soil farming, approximately 90% less  water and 40% faster harvesting time. Thinking that hydroponics, often described by the media as the solution to food demand for an overpopulated Earth, can fix population growth is not true. Current  literature presents several studies analysing the disavantages of landless methods. Among them: 

1)  Cumbersome chemistry management and nutrient distribution, 

2) High setup costs, electricity cost  and high greenhouse maintenance cost, 

3) Easy water-born contamination, pH unbalance and easy  spread of diseases and 

4) high labor and automation costs. 

In the MESOI project, we want to focus on the last two  problems.  We aim at developing a practical and reliable automatic method for the detection  of the most common plant diseases that can affect a typical cultivation. This solution focuses on an  autonomous analysis machine-vision method capable of detecting evolution stages of plants (presence of fruit) and presence of plant diseases (e.g. aphids).

 

The proposed solution can allow farmers to monitor the wealth of their crop and alert them when a  plant disease is present, pinpointing to where exactly the disease has spread reporting as well the  size of the infected area. The proposed system consists of a lightweight rover equipped with a machine learning algorithm constantly processing images of the crop.

Aquest projecte ha estat cofinançat per la Unió Europea a través del Fons Europeu de Desenvolupament Regional (FEDER) i compta amb el suport de la Secretaria d’Universitats i Recerca del Departament d’Empresa i Coneixement de la Generalitat de Catalunya.