Premium
Conventional digital cameras as a tool for assessing leaf area index and biomass for cereal breeding
Author(s) -
Casadesús Jaume,
Villegas Dolors
Publication year - 2014
Publication title -
journal of integrative plant biology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.734
H-Index - 83
eISSN - 1744-7909
pISSN - 1672-9072
DOI - 10.1111/jipb.12117
Subject(s) - overcast , biomass (ecology) , triticale , leaf area index , vegetation (pathology) , remote sensing , quadrat , environmental science , digital photography , index (typography) , sampling (signal processing) , pixel , agronomy , mathematics , photography , sky , geography , botany , biology , computer science , meteorology , medicine , art , pathology , shrub , visual arts , computer vision , filter (signal processing) , world wide web
Affordable and easy‐to‐use methods for assessing biomass and leaf area index (LAI) would be of interest in most breeding programs. Here, we describe the evaluation of a protocol for photographic sampling and image analysis aimed at providing low‐labor yet robust indicators of biomass and LAI. In this trial, two genotypes of triticale, two of bread wheat, and four of tritordeum were studied. At six dates during the growing cycle, biomass and LAI were measured destructively, and digital photography was taken on the same dates. Several vegetation indices were calculated from each image. The results showed that repeatable and consistent values of the indices were obtained in consecutive photographic samplings on the same plots. The photographic indices were highly correlated with the destructive measurements, though the magnitude of the correlation was lower after anthesis. This work shows that photographic assessment of biomass and LAI can be fast, affordable, have good repeatability, and can be used under bright and overcast skies. A practical vegetation index derived from pictures is the fraction of green pixels over the total pixels of the image, and as it shows good correlations with all biomass variables, is the most robust to lighting conditions and has easy interpretation.