Back to Search Start Over

Assessing the health and yield of argan trees in Morocco’s unique ecosystem: a multispectral and machine learning approach.

Authors :
Saddik, Amine
Hssaisoune, Mohammed
Belaqziz, Salwa
Labbaci, Adnane
Tairi, Abdellaali
Meskour, Brahim
Bouchaou, Lhoussaine
Source :
International Journal of Remote Sensing. Oct2024, p1-29. 29p. 10 Illustrations.
Publication Year :
2024

Abstract

This research is focused on developing an AI model that utilizes multispectral camera data from the Souss-Massa region. The model aims to estimate the yield and identify a range of diseases in Argan trees through meticulous on-field investigations. Initially, the work is focused on understanding the resistance of Argan plants against different diseases, based on non-irrigated and irrigated Argan trees as well as planted ones. The results show that disease resistance is high in the case of non-irrigated Argan trees and low in the case of irrigated trees. In addition, we conducted a detailed study of the Argan trees to provide a comprehensive view of the plant’s health. Utilizing machine-learning techniques, the yield estimation model suggests that it is possible to achieve up to 97% accuracy in yield estimation, processing data at an impressive rate of 33 images per second. After careful consideration and analysis, we have concluded that machine counting is the most suitable technique for Argan plants. Machine counting and disease detection offer high precision, fast and efficient data processing and cost-effectiveness. Additionally, it is less labour-intensive and can be easily integrated into the existing production process. [ABSTRACT FROM AUTHOR]

Details

Language :
English
ISSN :
01431161
Database :
Academic Search Index
Journal :
International Journal of Remote Sensing
Publication Type :
Academic Journal
Accession number :
180066602
Full Text :
https://doi.org/10.1080/01431161.2024.2407560