Miraki, M., Sohrabi, H., Fatehi, P. (2022). Citrus trees identification and trees stress detection based on spectral data derived from UAVs. , 1(1), 27-40. doi: 10.22092/rhsj.2022.127815
Mojdeh Miraki; Hormoz Sohrabi; Parviz Fatehi. "Citrus trees identification and trees stress detection based on spectral data derived from UAVs". , 1, 1, 2022, 27-40. doi: 10.22092/rhsj.2022.127815
Miraki, M., Sohrabi, H., Fatehi, P. (2022). 'Citrus trees identification and trees stress detection based on spectral data derived from UAVs', , 1(1), pp. 27-40. doi: 10.22092/rhsj.2022.127815
Miraki, M., Sohrabi, H., Fatehi, P. Citrus trees identification and trees stress detection based on spectral data derived from UAVs. , 2022; 1(1): 27-40. doi: 10.22092/rhsj.2022.127815

Citrus trees identification and trees stress detection based on spectral data derived from UAVs

پژوهش در علوم باغبانی
Volume 1, Issue 1, September 2022, Pages 27-40    PDF (964.87 K)
Document Type: Research
DOI: 10.22092/rhsj.2022.127815
Authors
Mojdeh Miraki1; Hormoz Sohrabi* 1; Parviz Fatehi2
1Department of Forestry, Faculty of Natural Resources, Tarbiat Modares University
2Department of Forestry and Forest Economics, Faculty of Natural Resources, University of Tehran, Karaj, Iran
Abstract
One of the requirements for achieving efficiency and maximum productivity is continuous monitoring and identifying of horticultural and agricultural products. Traditional plant monitoring and evaluation methods are time-consuming, labor-intensive, and costly. Unmanned aerial vehicles (UAVs) using real color imaging (red/green/blue) is a game-changer in horticultural and agriculture and an economically viable option for recognizing stress and disease. In this paper, the ability of UAV images was evaluated for identifying citrus trees and determining their health using a simple method. For this purpose, In June 2019, the study area was photographed and surveyed. The region growing algorithm was tested for a series of CHMs generated from point clouds, across a range of spatial resolutions. The highest overall accuracy for the individual tree crown delineation was achieved for a spatial resolution of 50 cm (F-score =0.63). In the next step, the orthomosaic was generated with a pixel size of 2.5 cm was generated by structure from motion algorithm. Then vegetation indices and bands obtained from orthomosaic and CIE L* a* b* color space were used as input data in a random forest classification algorithm. We classified the trees into 2 classes: health trees and unhealthy trees; then the random forest algorithm was applied using R software. The classification accuracy for identified trees was performed using 10-fold cross-validation. The classification resulted in overall accuracies of 69%; that display the effectiveness of UAVs to inform stakeholders.
Keywords
UAV"; Color space"; Stress"; Random forest algorithm"; "; Region-growing algorithm"
References
Statistics
Article View: 6,370
PDF Download: 618


counter
Journal management system. designed by sinaweb