Development of precision methods for remote monitoring of weeds

Roman Danilov,Mariya Ivanisova,Roman Zelensky,Artyom Ponomarev,Kseniya Gasiyan

doi:10.1051/bioconf/20202100003

Roman Danilov, Mariya Ivanisova + Show 3 more

Open Access

https://doi.org/10.1051/bioconf/20202100003

Copy DOI

Abstract

As a result of hyperspectral measurements carried out on the test plots of the All-Russian Research Institute of Biological Plant Protection, we obtained the reflection spectra of individual cultivated and weed species. Along with the ground-based measurements, trials of the GEOSCAN drone equipped with a Gamaya hyperspectral camera were organized. We developed a database of spectral characteristics of weeds. The change in the shape of the spectral signatures of the reflectivity of plant objects depending on their actual condition, taken into account during field surveys, was analyzed. An analysis of the spectral curves showed that the results of field spectrometry are in good agreement with aerial survey data.

Highlights

Weed plants are components of secondary habitats plant communities and essentially remain wild-growing plants, the distribution of which is caused, first of all, by natural factors, the main of which is the heat and moisture supply of the growing area, rather than being attached to a specific agricultural crop [1, 2].To carry out real-time monitoring of agrocenoses conditions, it is necessary to improve methods for obtaining information on weeds in specific regions with reference to agroclimatic conditions [3, 4]
On the experimental fields of scientific crop rotation of the AllRussian Research Institute of Biological Plant Protection, we developed test plots with different species composition of cultivated and weed plants that form the basis of the agrocenoses of Krasnodar Krai [16]
The objects of the research were crops of sunflower, corn, soybeans and alfalfa, to different extents weeded with the following species: ragweed (Ambrosia artemisiifolia L.), Californian burweed (Xanthium strumarium L.), black foxtail (Alopecurus myosuroides Huds.) and wild oat (Avena fatua L.)

Summary

Introduction

Weed plants are components of secondary habitats plant communities and essentially remain wild-growing plants, the distribution of which is caused, first of all, by natural factors, the main of which is the heat and moisture supply of the growing area, rather than being attached to a specific agricultural crop [1, 2].To carry out real-time monitoring of agrocenoses conditions, it is necessary to improve methods for obtaining information on weeds in specific regions with reference to agroclimatic conditions [3, 4]. Weed plants are components of secondary habitats plant communities and essentially remain wild-growing plants, the distribution of which is caused, first of all, by natural factors, the main of which is the heat and moisture supply of the growing area, rather than being attached to a specific agricultural crop [1, 2]. That's why it's really important to develop fundamentally new diagnostic tools and operational control of especially dangerous harmful species that can cause an emergency food situation [5,6,7]. The Russian experience in the use of UAVs in agriculture today mainly comes down to determining the area of agricultural land, the general economic assessment of crops conditions and the determination of the NDVI index (Normalized Difference Vegetation Index) [11,12,13]. There is a lack of proposals for the use of modern technical tools for diagnosing the species composition and prognostic models of development, and the spread of harmful objects

Objectives

Methods

Results

Conclusion