Aerial images and machine learning methods to emulate the late blight severity in potato crops
Assessment of Phytophthora infestans’ incidence and severity are frequently performed based on visual crop inspection, which is a labor-intensive task prone to errors associated with its subjectivity. Therefore, alternative methods to relate disease incidence and severity with changes in crop traits are of great interest. Optical imagery in the visible and near-infrared (VisNIR) can detect changes in crop traits caused by pathogen development. In addition, Unmanned Aerial Vehicles (UAV) with cameras on board have flexible data collection capabilities allowing adjustments considering the trade-off between data throughput and its resolution. This work presents a quantitative prediction of the severity of the disease caused by Phytophthora infestans in potato crops using image processing and machine learning (ML) algorithms such as Random Forests (RF) and Extreme Gradient Boost (XGBoost). The ML algorithms were trained using datasets from multispectral data captured at the canopy level with a UAV carrying a multispectral camera. The results indicate that RF and XGBoost using 11 classes with 18 bands, including vegetation indexes and band features, can predict late blight severity on potato crops with an acceptable accuracy of 81.02% for RF and 74.19% for RF XGBoost.
| Main Authors: | , , , , , |
|---|---|
| Format: | Report biblioteca |
| Language: | English |
| Published: |
International Potato Center
2022-12
|
| Subjects: | potatoes, phytophthora infestans, |
| Online Access: | https://hdl.handle.net/10568/128499 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|