Related Experiment Video
Updated: Oct 9, 2026
High Throughput Image-Based Phenotyping for Determining Morphological and Physiological Responses to Single and Combined Stresses in Potato
Published on: June 7, 2024
Potato LAI estimation by enhancing spectral features of reflectance and fluorescence to improve application potential
Mingjia Liu1, Guohui Liu1, Changle Guo1
1Key Laboratory of Smart Agriculture System Integration, Ministry of Education, China Agricultural University, Beijing 100083, China.
Abstract:
Accurate leaf area index (LAI) estimation of potato crops is of great importance for precision management in potato cultivation. Vegetation indices (VIs) derived from spectral reflectance are widely used for LAI estimation but suffer from saturation under high LAI conditions, limiting accuracy and robustness across varieties. Solar-induced chlorophyll fluorescence in the near-infrared region (SIFNIR), closely linked to canopy photosynthetic activity and light absorption, provides complementary information for LAI estimation. However, heterogeneous light radiation and structural differences among canopies interfere with SIFNIR extraction. To address these challenges, we propose an approach that enhances and integrates spectral reflectance and fluorescence features to improve LAI estimation across 40 potato varieties. Seven VIs were calculated and their saturation effects were discussed. SIFNIR was extracted based on Three-band Fraunhofer line discriminator (3FLD) method and normalized by absorbed photosynthetically active radiation (APAR) and canopy coverage (Cov) to reduce radiative and structural effects. Potato LAI estimation models of different varieties were constructed using the Random Forest (RF) method. The results showed that (1) NLI and MSAVI were reliable VIs for assessing potato canopy growth. (2) APAR and Cov showed stronger relationships with LAI, and SIFNIR normalized by these factors could reduce interference from canopy structure. (3) The RF model of reflectance and fluorescence feature fusion achieved the best performance, with prediction set Rp2 of 0.68, 0.70, 0.68, 0.70, and RMSEp of 0.50, 0.45, 0.50, 0.45 for all varieties, early season, mid-season and late season varieties, respectively. These results demonstrate the application potential of reflectance and fluorescence fusion to improve potato LAI estimation, and provide support for precise management in potato cultivation.
