High-throughput Mapping of Leaf Pigment Content from Multispectral Data

Accurate and timely assessment of the physiological status of plants is crucial for resource optimization and minimizing environmental impact in ecological monitoring and agriculture.This study introduces a novel technique for mapping pigment concentrations in plant leaves using multispectral imaging.The key innovation of the proposed approach is a multispectral camera equipped with reconfigurable spectral filters, enabling application-specific data acquisition.Combined with a data processing protocol, the system achieves high measurement accuracy of leaf pigment content even with limited spectral channels.The technique includes camera calibration, spatiospectral correction, spectral index selection, and nonlinear regression to develop empirical models for estimating chlorophyll concentration.Validation experiments on proximal sensing of lettuce (Lactuca sativa) demonstrated that the ratio of Modified Chlorophyll Absorption Ratio Index to Optimized Soil Adjusted Vegetation Index achieved the best performance, with an R² of 0.81, an RMSE of 0.3 mg/L, and relative error of less than 20%.Limitations include residual shading and interpolation artifacts, which suggest opportunities for further improvement.The findings highlight high potential of this technique for noninvasive and high-throughput monitoring of plant health in both ecological and agricultural contexts.