Treffer: Classification of yellow rust of wheat from Sentinel-2 satellite imagery using deep learning artificial neural network.

The yellow rust of wheat is a critical biotic stress causing the reduction in crop yield in different parts of the world. We hypothesized that identifying yellow rust from Sentinel-2 satellite imagery using a deep learning-based artificial neural network (ANN) classification model would be an efficient monitoring technique of the disease in the parts of Indian Punjab (Jalandhar/Kapurthala and Rupnagar). Six spectral indices (NDVI, NDWI, MCARI, Cl-Red Edge, NDMI, S2REP) were computed from the cloud-free Sentinel-2 data (9 February 2020 for Jalandhar/Kapurthala and 29 January 2020 for Rupnagar), and the random points were generated in disease and healthy wheat growing fields of the area. The pixel values of the spectral indices were extracted using the vector points, and this data was fed to a deep learning ANN classifier. The implementation of these models was performed in Python using Jupyter notebooks and standard python libraries like Keras, NumPy, and Pandas. The ANN model for classifications of yellow rust of wheat was different for both areas due to time differences of Sentinel-2, and the data of one area could not be classified using the trained model of the other area. The accuracy and F1 score were > 0.90 for both areas. These results suggest that Sentinel-2 data can be used to monitor the yellow rust of wheat. [ABSTRACT FROM AUTHOR]

Copyright of Arabian Journal of Geosciences is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)