The method of determining the amount of yield based on the results of remote sensing obtained using UAV on the example of wheat

The methods of contact monitoring of conditions of plants are not adapted for large-scale industrial field studies. Satellite monitoring was primarily created to objectively evaluate the prospects of harvesting. The introduction of UAVs has allowed to improve the accuracy of the results to some extent, due to the high resolution of the sensor equipment. In most cases, the prediction is based on standard NDVI stress indexes. However, these studies do not provide information on quantitative yield indicators that farms need in order to optimize logistics and use harvesting equipment. The development of a methodology for interpreting the results of remote monitoring in the form of values of vegetation (stress) indices in the expected volumes of the crop was the goal of the work. Researches were conducted in 2019 in the Kiev region on the production area of winter wheat crops in the separate subdivision of NULES of Ukraine "Agronomic Research Station". Wheat was harvested using a John Deere 9670STS combine. Data were recorded every second during the combine operation. In this case, more than 14,000 separate sections were obtained with a header width of 9 meters. The data were processed using software product Trimble (R) Farm Works (R) Office and exported as Microsoft Excel spreadsheets. A specialized Slantrange 3p spectral system was used for remote monitoring, which was mounted on the DJI Matrice 600 Pro industrial platform. Radio Frequency correction for changes in lighting system was carried out by the standard anti-aircraft sensor. The results were calculated and the stress index maps were constructed using the software of the SlantView system developer.On the basis of the experimental data obtained, was developed a technique to compare the objective harvest data obtained by the harvester in terms of grain count and UAV stress index values. It is suggested to use a stress index, which makes it possible to predict the harvest (the map of the stress indexes was built 2 months before the harvest). Above mentioned allows the farm to optimize the use of ground equipment.