Diagnosis decline in citrus using multispectral camera-equipped unmanned aerial system
Author(s) -
Erfan Seidipoor,
Farhad Samadzadegan,
Farzaneh Dadrass Javan,
Omid Askari
Publication year - 2018
Publication title -
journal of geospatial information technology
Language(s) - English
Resource type - Journals
eISSN - 2538-418X
pISSN - 2008-9635
DOI - 10.29252/jgit.6.3.39
Subject(s) - multispectral image , remote sensing , usable , computer science , vegetation index , flexibility (engineering) , environmental science , normalized difference vegetation index , geography , climate change , mathematics , ecology , statistics , biology , world wide web
Today getting access to the monitoring systems with high accuracy in order to produce healthy agricultural products has been raised though the following effective factors are of high importance as well; prevent pest risk and reduce the coast raised from unmanned aerial systems according to low cost, low weight, possibility of taking images in cloudy weather under the cover of clouds, flexibility in time and high spatial resolution imaging which are usable in obtaining remote sensing data and considerable interest. In this regard, the use of these systems shows high potential in obtaining required data in monitoring products and provided control. In this paper, multi-spectral sensor mounted on an unmanned airborne platform in the health status of the regional citrus trees infected with the decline disease in the Fars province investigated. Using data by this system, classified images of studied area based on vegetation index and distinctions between healthy and infected trees, use of SVM are provided. Collected ground data shows the feasibility of these methods in tree’s health status diagnosis. About 90 percent overall accuracy was achieved in the trees classification.
Accelerating Research
Robert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom
Address
John Eccles HouseRobert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom