Remote Sensing, Monitoring and Classification of Agricultural Land Using Artificial Neural Networks

EasyChair Preprint no. 2812

Abstract

The results of an analytical review of the use of artificial neural networks (ANN) in agricultural production are presented. Results of remote sensing of agricultural land in the form of color images can be obtained using unmanned aerial vehicles (UAV). UAV can be used as aerial robots that perform aerial photography functions, transport technological components such as plant protection products, and perform other similar functions. Other functional equipment can be installed on the aircraft: thermal imagers, multispectral and IR cameras, etc., using the data obtained from the UAV, you can create an orthophotoplane or a 3D model of the area, make a map of heights, determine the state of fields, crops and determine their vegetation indices NDVI. A classification of UAV application areas in agricultural production is proposed, which involves ordering the UAV application areas in agriculture. To obtain and process remote sensing data from UAVs in different parts of the spectrum, a conceptual model of the software package has been developed. The software package is intended for receiving and processing the results of monitoring and subsequent analysis of the set of calculated vegetation indices. The main research tasks solved by the software developer and determining its structure are formulated. To predict productivity, a method of applying the results of aerial photography in combination with experimental data on the biological development of agricultural crops has been developed. For practical use of the developed methodology, a database for each culture is formed. The results obtained from UAV are used to construct regression and matrix mathematical models of the relationship between optical-spectral characteristics and crop yields.

Keyphrases: Agricultural crops, Artificial Neural Networks, monitoring, Productivity, Unmanned Aerial Vehicles

@Booklet{EasyChair:2812,
  author = {Aleksey Rogachev and Elena Melikhova},
  title = {Remote Sensing, Monitoring and Classification of Agricultural Land Using Artificial Neural Networks},
  howpublished = {EasyChair Preprint no. 2812},

  year = {EasyChair, 2020}}