Analysis of polarization characteristics of plant canopies using ground-based remote sensing measurements

Sid'ko, A. F.; Pisman, T.I.; Shevyrnogov, A.

doi:10.1016/j.jqsrt.2014.03.031

Cited by 4 publications

(1 citation statement)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These works concern the detection and diagnosis of diseases/pests (e.g., Zhang et al 2003;Bauriegel et al 2011), plant growth monitoring and yield prediction (e.g., Freeman et al 2006;Pittman et al 2015), diagnostics of nutrient deficiency and stress conditions (e.g., Chaerle & Van der Straeten 2000;Tilling et al 2007). On the other hand, much research is devoted to the identification and functioning of plants (e.g., Hestira et al 2008;Ustin & Gamon 2010;Adam et al 2010;Sid'ko et al 2014).…”

Section: Introductionmentioning

confidence: 99%

Hyperspectral imaging in assessing the condition of plants: strengths and weaknesses

Dominiak-Świgoń

Olejniczak

Nowak

et al. 2019

Biodiversity Research and Conservation

View full text Add to dashboard Cite

Hyperspectral remote sensing of plants is widely used in agriculture and forestry. Fast, large-area monitoring is applied, among others, in detecting and diagnosing diseases, stress conditions or predicting the yields. Using available tools to increase the yields of most important crop plants (wheat, rice, corn) without posing threat to food security is essential in the situation of current climate changes. Spectral plant indices are associated with biochemical and biophysical plant characteristics. Using the plant spectral properties (mainly chlorophyll red light absorption and near-infrared range light reflectance in leaf intercellular spaces), it is possible to estimate plant condition, water and carotenoid contents or detect disease. More and more often, based on commonly used hyperspectral vegetation indices, new, more sensitive indices are introduced. Furthermore, to facilitate data processing, artificial intelligence is employed, i.e., neural networks and deep convolutional neural networks. It is important in ecological research to carry out long-term observations and measurements of organisms throughout their lifespan. A non-invasive, quick method ensures that it may be used many times and at each stage of plant development.

show abstract