Published in Scientific Papers. Series "Management, Economic Engineering in Agriculture and rural development", Vol. 23 ISSUE 4
Written by Pavlo LYKHOVYD, Sergiy LAVRENKO, Nataliia LAVRENKO

The purpose of the article was to describe the outcomes of an in-field evaluation of the NDVI Converter application, which was created to assess the normalized difference vegetation index (NDVI) using estimates of the percentage of green canopy cover. The study was carried out in 2023 on fields of winter wheat, winter barley, and winter oilseed rape in the phenological stages BBCH 21–32 and BBCH 18–39, respectively. The fraction of green canopy cover was estimated using the application Canopeo. By comparing the mean absolute percentage error and Pearson's correlation coefficient to the actual values of the spatial vegetation index, as determined by the platform OneSoil, the quality of the vegetation index evaluation was assessed. Thus, it has been established that NDVI Converter offers accurate vegetation index assessment for cereal crops, with mean absolute percentage error 16.23% and Pearson’s correlation coefficient within 0.99, and reasonable quality for oilseed rape (statistical indicators of 46.61% and 0.99, respectively). After the adjustment, the accuracy score of the NDVI Converter increased up to 27.62% for oilseed rape, and 8.71% for winter cereals. Therefore, NDVI Converter could be recommended for practical use in case the spatial vegetation index is not provided by satellite imagery services.

[Read full article] [Citation]