Methodology and software for atmospheric correction of images obtained with unmanned aircraft to allow the safe location of vegetation

Abstract

Unmanned aerial vehicles (UAV) provide a way to obtain images with high spatial resolution with relatively simple control of the vehicle and economic efficiency. Over the past 10 years, UAV measurement technology has changed significantly and has become popular due to its versatility and ease of use in commercial and scientific research. It should be noted that the methods of processing one-time received images allow solving various scientific and practical problems, however, during monitoring, when the dynamics of the state of objects (surface types) of the same territory is studied, the methodology and software remain questionable. One of the problems that hinder the study, for example, the dynamics of the state of objects on an agricultural field, is the radiometric accuracy of the obtained UAV images. In practice, the radiometric accuracy of measurements allows considering various lighting conditions at different periods of the day and year, by relevant types of digital sensors (digital cameras) used. The study is aimed at enabling a better flexibility when studying the dynamics of the state of objects located on the same territory at different times by taking into account the radiometric errors of UAV images when mapping the state of vegetation. The research methodology uses the real images of the UAV obtained from June to August 2019 for an agricultural field with winter wheat.