A Comparative Study of Colorimetry and Reflectance Spectroscopy for Estimating Pigment Content in Plant Leaves

In this study, we compared colorimetry and reflectance spectroscopy to quantify pigment content in plant leaves. We selected leaf samples from Hibiscus rosa-sinensis, Syngonium, and Lactuca sativa at various growth stages and colors to create a diverse dataset. Using a spectrometer with uniform spectral and brightness illumination, we measured and processed reflectance spectra, calculating chromaticity coordinates in the CIE 1931 xyY and CIELAB color spaces, along with ten vegetation indices sensitive to chlorophyll content. Additionally, we performed a spectrophotometric analysis of leaf extracts to obtain optical density spectra for estimating chlorophyll concentration using the traditional method. We developed regression models to analyze relationships between chlorophyll content and both vegetation indices and chromaticity coordinates. The MCARI/OSAVI vegetation index demonstrated the highest predictive accuracy ($\mathrm{R}^{2}=0.91, \text{RE}=17 \%$), whereas colorimetric parameters in xyY and Lab spaces showed weaker correlations ($\mathrm{R}^{2} \leq 0.53$, RE up to 36 %). Despite its limited predictive power for chlorophyll content, colorimetry offers advantages such as simplicity, low cost, and rapid measurement, making it useful for preliminary assessments and applications where high precision is not critical. Moreover, this study initiated the development of a spectral database for plant condition monitoring, incorporating spectral channels in both visible and near-infrared ranges. This approach accounts for species-specific physiological differences and supports enhanced assessment of plant physiological status, which is crucial for precision agriculture, horticulture, and plant phenotyping.