Naslov
Detection and evaluation of environmental stress in winter wheat using proximal sensing methods

Autori
Skendžić, Sandra ; Zovko, Monika ; Lešić, Vinko ; Maričević, Marko ; Pajač Živković, Ivana ; Lemić, Darija

Vrsta, podvrsta i kategorija rada
Sažeci sa skupova, sažetak, znanstveni

Izvornik
Digital Technologies in Agriculture, Book of Abstracts / Lončarić, Zdenko ; Jović, Jurica - Osijek : 1, 2022, 23-23

ISBN
978-953-8421-03-7

Skup
1st International Symposium on Digital Technologies in Agriculture (ISDTA 2022) ; 1st Satellite Workshop Digital Agriculture in Rural Area (DIGITAGRA 2022)

Mjesto i datum
Osijek, Hrvatska, 06.12.2022. - 08.12.2022

Vrsta sudjelovanja
Poster

Vrsta recenzije
Recenziran

Ključne riječi
climate change ; winter wheat ; reflectance ; remote sensing ; environmental stress

Sažetak
Climate change has significant impacts on winter wheat (Triticum aestivum L.) production through the occurrence of various environmental stress parameters. In this sense, drought stress and pest infestation, especially in sensitive phenophases, are the main threats to wheat production stability and final yield loss. These stress factors lead to changes in the physiological response of winter wheat plants and can be monitored by various multi- and hyperspectral remote sensing systems (satellite, airborne, and proximal). One of the best known and most accurate methods for monitoring plant stress is proximal remote sensing. As plant responses to environmental stress are numerous and complex, their physiological effects affect the signal in different regions of the electromagnetic spectrum, such as visible (VIS), near infrared (NIR), and shortwave infrared (SWIR). Through their feeding, pests cause changes in plant pigments and cell structure, resulting in variations in VIS and NIR reflectance, while plant water content has been associated with variations in certain bands in the SWIR region of the spectrum. Picture 1. Workflow for detection and evaluation of environmental stress in crops using proximal sensing methods for precise agricultural management decisions The availability of hyperspectral data has led to the definition of vegetation indices (VI), which, by combining the values of different regions of spectrum with plant (or soil) characteristics, can quantify many agronomic variables and estimate the severity of plant stress. In addition to VI, the total spectral signature (reflectance) or its derivatives and principal components (PC) are often used to assess plant condition. Chemometric (multivariate) and machine learning models are commonly used to determine the relationship between groundthruth parameters (water content in plants/ soil, level of pest infestation, etc.) and leaf/canopy spectral characteristics. These include principal component analysis (PCA), partial least squares regression (PLSR) and principal component regression (PCR), artificial neural networks (ANN) and support vector machines (SVM).

Izvorni jezik
Engleski

Znanstvena područja
Interdisciplinarne tehničke znanosti, Poljoprivreda (agronomija), Interdisciplinarne biotehničke znanosti

POVEZANOST RADA