TY - JOUR
KW - Pest Management
KW - Computer vision
KW - Machine Learning
KW - Artificial Intelligence
KW - Blockchain
KW - Farming
KW - Greenhouses
AU - Miguel Angel Lopez
AU - Juan Manuel Lombardo
AU - Mabel López
AU - David Álvarez
AU - Susana Velasco
AU - Sara Terrón
AB - The appearance of pests is one of the major problems that exist in the growth of crops, as they can damage the production if the appropriate measures are not taken. Within the framework of the Integrated Pest Management strategy (IPM), early detection of pests is an essential step in order to provide the most appropriate treatment and avoid losses. This paper proposes the architecture of a system intensive farming in greenhouses featuring the ability to detect environmental variations that may favour the appearance of pests. This system can suggest a plan or treatment that will help mitigate the effects that the identified pest would produce otherwise. Furthermore, the system will learn from the actions carried out by the humans throughout the different stages of crop growing and will add it as knowledge for the prediction of future actions. The data collected from sensors, through computer vision, or the experiences provided by the experts, along with the historical data related to the crop, will allow for the development of a model that contrasts the predictions of the actions that could be implemented with those already performed by technicians. Within the technological ecosystems in which the Integrated Pest Management systems develop their action, traceability models must be incorporated. This will guarantee that the data used for the exploitation of the information and, therefore for the parameterization of the predictive models, are adequate. Thus, the integration of blockchain technologies is considered key to provide them with security and confidence.
IS - Special Issue on Artificial Intelligence and Blockchain
M1 - 3
N2 - The appearance of pests is one of the major problems that exist in the growth of crops, as they can damage the production if the appropriate measures are not taken. Within the framework of the Integrated Pest Management strategy (IPM), early detection of pests is an essential step in order to provide the most appropriate treatment and avoid losses. This paper proposes the architecture of a system intensive farming in greenhouses featuring the ability to detect environmental variations that may favour the appearance of pests. This system can suggest a plan or treatment that will help mitigate the effects that the identified pest would produce otherwise. Furthermore, the system will learn from the actions carried out by the humans throughout the different stages of crop growing and will add it as knowledge for the prediction of future actions. The data collected from sensors, through computer vision, or the experiences provided by the experts, along with the historical data related to the crop, will allow for the development of a model that contrasts the predictions of the actions that could be implemented with those already performed by technicians. Within the technological ecosystems in which the Integrated Pest Management systems develop their action, traceability models must be incorporated. This will guarantee that the data used for the exploitation of the information and, therefore for the parameterization of the predictive models, are adequate. Thus, the integration of blockchain technologies is considered key to provide them with security and confidence.
PY - 2020
SP - 47
EP - 54
T2 - International Journal of Interactive Multimedia and Artificial Intelligence
TI - Traceable Ecosystem and Strategic Framework for the Creation of an Integrated Pest Management System for Intensive Farming
UR - https://www.ijimai.org/journal/sites/default/files/2020-08/ijimai_6_3_6.pdf
VL - 6
SN - 1989-1660
ER -