Abstract
Leaf area index (LAI) is a critical indicator of canopy architecture and physiological performance, serving as a key parameter for crop growth monitoring and management. Although UAV multispectral imagery provides rich spectral and spatial information, the limitations of single texture features for LAI estimation still require further exploration. To address this issue, this study developed a multi-source feature fusion framework that integrates vegetation indices (VIs), texture features (TFs), and texture indices (TIs) within a stacked ensemble approach combining Partial Least Squares Regression (PLSR) with Support Vector Machine (SVM), Random Forest (RF), and Gradient Boosting Decision Tree (GBDT) algorithms to estimate maize LAI.A field experiment was conducted under three planting densities (42,000, 63,000, and 84,000 plants ha−1) and four nitrogen rates (0, 80, 160, 240 kg N ha−1) to assess the potential of UAV-based multispectral imagery for maize LAI estimation. The results show that when using partial least squares regression (PLSR) combined with RF, SVM and GBDT to estimate maize LAI, the R2 values are 0.653, 0.697 and 0.634, and the RMSE is 0.650, 0.608 and 0.668, respectively, when only vegetation indices (VIs) is used as input. After texture features (TFs) incorporation, the R2 increases to 0.717, 0.794, and 0.801, and the RMSE decreases to 0.587, 0.500, and 0.492. Further inclusion of the texture indices (TIs) raises the R2 to 0.789, 0.804, and 0.844, with RMSE of 0.506, 0.489, and 0.436, respectively. Independent test set validation under contrasting conditions confirmed that our multi-model fusion framework (PLSR+GBDT) with multi-source feature fusion (VIs+TFs+TIs) effectively estimated LAI, achieving an R2 of 0.859 and 0.794. These results demonstrate that multi-source feature integration via machine learning enables robust and accurate estimation of maize LAI, providing a valuable tool for precision agriculture and crop growth monitoring.
Abstract from the paper’s OpenAlex record; open copy licensed cc-by. Read at the source ↗
The text the model read, with the source record’s inline markup (italics, sub- and superscripts) rendered and its character entities decoded; quote offsets refer to the raw text.
Claims 19
Claim · model-written summary
Using only VIs, PLSR combined with RF estimated maize LAI with an R2 of 0.653.
- Treatment
- only vegetation indices (VIs)e1
- Outcome
- R2e1
- Value
- 0.653e1
- Qualifier
- partial least squares regression (PLSR) combined with RF, SVM and GBDTe0
Zea maysOrganisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“partial least squares regression (PLSR) combined with RF, SVM and GBDT to estimate maize LAI”
e1“the R2 values are 0.653, 0.697 and 0.634, and the RMSE is 0.650, 0.608 and 0.668, respectively, when only vegetation indices (VIs) is used as input”
t0“Estimating Maize Leaf Area Index Using Multi-Source Features Derived from UAV Multispectral Imagery and Machine Learning Models”
Claim · model-written summary
Using only VIs, PLSR combined with SVM estimated maize LAI with an R2 of 0.697.
- Treatment
- only vegetation indices (VIs)e1
- Outcome
- R2e1
- Value
- 0.697e1
- Qualifier
- partial least squares regression (PLSR) combined with RF, SVM and GBDTe0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“partial least squares regression (PLSR) combined with RF, SVM and GBDT to estimate maize LAI”
e1“the R2 values are 0.653, 0.697 and 0.634, and the RMSE is 0.650, 0.608 and 0.668, respectively, when only vegetation indices (VIs) is used as input”
t0“Estimating Maize Leaf Area Index Using Multi-Source Features Derived from UAV Multispectral Imagery and Machine Learning Models”
Claim · model-written summary
Using only VIs, PLSR combined with GBDT estimated maize LAI with an R2 of 0.634.
- Treatment
- only vegetation indices (VIs)e1
- Outcome
- R2e1
- Value
- 0.634e1
- Qualifier
- partial least squares regression (PLSR) combined with RF, SVM and GBDTe0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“partial least squares regression (PLSR) combined with RF, SVM and GBDT to estimate maize LAI”
e1“the R2 values are 0.653, 0.697 and 0.634, and the RMSE is 0.650, 0.608 and 0.668, respectively, when only vegetation indices (VIs) is used as input”
t0“Estimating Maize Leaf Area Index Using Multi-Source Features Derived from UAV Multispectral Imagery and Machine Learning Models”
Claim · model-written summary
Using only VIs, PLSR combined with RF gave an RMSE of 0.650 for maize LAI.
- Treatment
- only vegetation indices (VIs)e1
- Outcome
- RMSEe1
- Value
- 0.650e1
- Qualifier
- partial least squares regression (PLSR) combined with RF, SVM and GBDTe0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“partial least squares regression (PLSR) combined with RF, SVM and GBDT to estimate maize LAI”
e1“the R2 values are 0.653, 0.697 and 0.634, and the RMSE is 0.650, 0.608 and 0.668, respectively, when only vegetation indices (VIs) is used as input”
t0“Estimating Maize Leaf Area Index Using Multi-Source Features Derived from UAV Multispectral Imagery and Machine Learning Models”
Claim · model-written summary
Using only VIs, PLSR combined with SVM gave an RMSE of 0.608 for maize LAI.
- Treatment
- only vegetation indices (VIs)e1
- Outcome
- RMSEe1
- Value
- 0.608e1
- Qualifier
- partial least squares regression (PLSR) combined with RF, SVM and GBDTe0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“partial least squares regression (PLSR) combined with RF, SVM and GBDT to estimate maize LAI”
e1“the R2 values are 0.653, 0.697 and 0.634, and the RMSE is 0.650, 0.608 and 0.668, respectively, when only vegetation indices (VIs) is used as input”
t0“Estimating Maize Leaf Area Index Using Multi-Source Features Derived from UAV Multispectral Imagery and Machine Learning Models”
Claim · model-written summary
Using only VIs, PLSR combined with GBDT gave an RMSE of 0.668 for maize LAI.
- Treatment
- only vegetation indices (VIs)e1
- Outcome
- RMSEe1
- Value
- 0.668e1
- Qualifier
- partial least squares regression (PLSR) combined with RF, SVM and GBDTe0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“partial least squares regression (PLSR) combined with RF, SVM and GBDT to estimate maize LAI”
e1“the R2 values are 0.653, 0.697 and 0.634, and the RMSE is 0.650, 0.608 and 0.668, respectively, when only vegetation indices (VIs) is used as input”
t0“Estimating Maize Leaf Area Index Using Multi-Source Features Derived from UAV Multispectral Imagery and Machine Learning Models”
Claim · model-written summary
Adding texture features to VIs, PLSR combined with RF estimated maize LAI with an R2 of 0.717.
- Treatment
- texture features (TFs) incorporatione2
- Outcome
- R2e2
- Value
- 0.717e2
- Qualifier
- partial least squares regression (PLSR) combined with RF, SVM and GBDTe0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“partial least squares regression (PLSR) combined with RF, SVM and GBDT to estimate maize LAI”
e2“After texture features (TFs) incorporation, the R2 increases to 0.717, 0.794, and 0.801, and the RMSE decreases to 0.587, 0.500, and 0.492”
t0“Estimating Maize Leaf Area Index Using Multi-Source Features Derived from UAV Multispectral Imagery and Machine Learning Models”
Claim · model-written summary
Adding texture features to VIs, PLSR combined with SVM estimated maize LAI with an R2 of 0.794.
- Treatment
- texture features (TFs) incorporatione2
- Outcome
- R2e2
- Value
- 0.794e2
- Qualifier
- partial least squares regression (PLSR) combined with RF, SVM and GBDTe0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“partial least squares regression (PLSR) combined with RF, SVM and GBDT to estimate maize LAI”
e2“After texture features (TFs) incorporation, the R2 increases to 0.717, 0.794, and 0.801, and the RMSE decreases to 0.587, 0.500, and 0.492”
t0“Estimating Maize Leaf Area Index Using Multi-Source Features Derived from UAV Multispectral Imagery and Machine Learning Models”
Claim · model-written summary
Adding texture features to VIs, PLSR combined with GBDT estimated maize LAI with an R2 of 0.801.
- Treatment
- texture features (TFs) incorporatione2
- Outcome
- R2e2
- Value
- 0.801e2
- Qualifier
- partial least squares regression (PLSR) combined with RF, SVM and GBDTe0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“partial least squares regression (PLSR) combined with RF, SVM and GBDT to estimate maize LAI”
e2“After texture features (TFs) incorporation, the R2 increases to 0.717, 0.794, and 0.801, and the RMSE decreases to 0.587, 0.500, and 0.492”
t0“Estimating Maize Leaf Area Index Using Multi-Source Features Derived from UAV Multispectral Imagery and Machine Learning Models”
Claim · model-written summary
Adding texture features to VIs, PLSR combined with RF gave an RMSE of 0.587 for maize LAI.
- Treatment
- texture features (TFs) incorporatione2
- Outcome
- RMSEe2
- Value
- 0.587e2
- Qualifier
- partial least squares regression (PLSR) combined with RF, SVM and GBDTe0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“partial least squares regression (PLSR) combined with RF, SVM and GBDT to estimate maize LAI”
e2“After texture features (TFs) incorporation, the R2 increases to 0.717, 0.794, and 0.801, and the RMSE decreases to 0.587, 0.500, and 0.492”
t0“Estimating Maize Leaf Area Index Using Multi-Source Features Derived from UAV Multispectral Imagery and Machine Learning Models”
Claim · model-written summary
Adding texture features to VIs, PLSR combined with SVM gave an RMSE of 0.500 for maize LAI.
- Treatment
- texture features (TFs) incorporatione2
- Outcome
- RMSEe2
- Value
- 0.500e2
- Qualifier
- partial least squares regression (PLSR) combined with RF, SVM and GBDTe0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“partial least squares regression (PLSR) combined with RF, SVM and GBDT to estimate maize LAI”
e2“After texture features (TFs) incorporation, the R2 increases to 0.717, 0.794, and 0.801, and the RMSE decreases to 0.587, 0.500, and 0.492”
t0“Estimating Maize Leaf Area Index Using Multi-Source Features Derived from UAV Multispectral Imagery and Machine Learning Models”
Claim · model-written summary
Adding texture features to VIs, PLSR combined with GBDT gave an RMSE of 0.492 for maize LAI.
- Treatment
- texture features (TFs) incorporatione2
- Outcome
- RMSEe2
- Value
- 0.492e2
- Qualifier
- partial least squares regression (PLSR) combined with RF, SVM and GBDTe0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“partial least squares regression (PLSR) combined with RF, SVM and GBDT to estimate maize LAI”
e2“After texture features (TFs) incorporation, the R2 increases to 0.717, 0.794, and 0.801, and the RMSE decreases to 0.587, 0.500, and 0.492”
t0“Estimating Maize Leaf Area Index Using Multi-Source Features Derived from UAV Multispectral Imagery and Machine Learning Models”
Claim · model-written summary
Adding texture indices to VIs and TFs, PLSR combined with RF estimated maize LAI with an R2 of 0.789.
- Treatment
- texture indices (TIs)e3
- Outcome
- R2e3
- Value
- 0.789e3
- Qualifier
- partial least squares regression (PLSR) combined with RF, SVM and GBDTe0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“partial least squares regression (PLSR) combined with RF, SVM and GBDT to estimate maize LAI”
e3“Further inclusion of the texture indices (TIs) raises the R2 to 0.789, 0.804, and 0.844, with RMSE of 0.506, 0.489, and 0.436, respectively”
t0“Estimating Maize Leaf Area Index Using Multi-Source Features Derived from UAV Multispectral Imagery and Machine Learning Models”
Claim · model-written summary
Adding texture indices to VIs and TFs, PLSR combined with SVM estimated maize LAI with an R2 of 0.804.
- Treatment
- texture indices (TIs)e3
- Outcome
- R2e3
- Value
- 0.804e3
- Qualifier
- partial least squares regression (PLSR) combined with RF, SVM and GBDTe0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“partial least squares regression (PLSR) combined with RF, SVM and GBDT to estimate maize LAI”
e3“Further inclusion of the texture indices (TIs) raises the R2 to 0.789, 0.804, and 0.844, with RMSE of 0.506, 0.489, and 0.436, respectively”
t0“Estimating Maize Leaf Area Index Using Multi-Source Features Derived from UAV Multispectral Imagery and Machine Learning Models”
Claim · model-written summary
Adding texture indices to VIs and TFs, PLSR combined with GBDT estimated maize LAI with an R2 of 0.844.
- Treatment
- texture indices (TIs)e3
- Outcome
- R2e3
- Value
- 0.844e3
- Qualifier
- partial least squares regression (PLSR) combined with RF, SVM and GBDTe0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“partial least squares regression (PLSR) combined with RF, SVM and GBDT to estimate maize LAI”
e3“Further inclusion of the texture indices (TIs) raises the R2 to 0.789, 0.804, and 0.844, with RMSE of 0.506, 0.489, and 0.436, respectively”
t0“Estimating Maize Leaf Area Index Using Multi-Source Features Derived from UAV Multispectral Imagery and Machine Learning Models”
Claim · model-written summary
Adding texture indices to VIs and TFs, PLSR combined with RF gave an RMSE of 0.506 for maize LAI.
- Treatment
- texture indices (TIs)e3
- Outcome
- RMSEe3
- Value
- 0.506e3
- Qualifier
- partial least squares regression (PLSR) combined with RF, SVM and GBDTe0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“partial least squares regression (PLSR) combined with RF, SVM and GBDT to estimate maize LAI”
e3“Further inclusion of the texture indices (TIs) raises the R2 to 0.789, 0.804, and 0.844, with RMSE of 0.506, 0.489, and 0.436, respectively”
t0“Estimating Maize Leaf Area Index Using Multi-Source Features Derived from UAV Multispectral Imagery and Machine Learning Models”
Claim · model-written summary
Adding texture indices to VIs and TFs, PLSR combined with SVM gave an RMSE of 0.489 for maize LAI.
- Treatment
- texture indices (TIs)e3
- Outcome
- RMSEe3
- Value
- 0.489e3
- Qualifier
- partial least squares regression (PLSR) combined with RF, SVM and GBDTe0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“partial least squares regression (PLSR) combined with RF, SVM and GBDT to estimate maize LAI”
e3“Further inclusion of the texture indices (TIs) raises the R2 to 0.789, 0.804, and 0.844, with RMSE of 0.506, 0.489, and 0.436, respectively”
t0“Estimating Maize Leaf Area Index Using Multi-Source Features Derived from UAV Multispectral Imagery and Machine Learning Models”
Claim · model-written summary
Adding texture indices to VIs and TFs, PLSR combined with GBDT gave an RMSE of 0.436 for maize LAI.
- Treatment
- texture indices (TIs)e3
- Outcome
- RMSEe3
- Value
- 0.436e3
- Qualifier
- partial least squares regression (PLSR) combined with RF, SVM and GBDTe0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“partial least squares regression (PLSR) combined with RF, SVM and GBDT to estimate maize LAI”
e3“Further inclusion of the texture indices (TIs) raises the R2 to 0.789, 0.804, and 0.844, with RMSE of 0.506, 0.489, and 0.436, respectively”
t0“Estimating Maize Leaf Area Index Using Multi-Source Features Derived from UAV Multispectral Imagery and Machine Learning Models”
Claim · model-written summary
The multi-model fusion framework PLSR+GBDT with VIs, TFs and TIs estimated maize LAI with an R2 of 0.859 and 0.794 in independent test sets under contrasting conditions.
- Treatment
- multi-source feature fusion (VIs+TFs+TIs)e4
- Outcome
- R2e4
- Value
- 0.859 and 0.794e4
- Qualifier
- Independent test set validation under contrasting conditionse4
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e4“Independent test set validation under contrasting conditions confirmed that our multi-model fusion framework (PLSR+GBDT) with multi-source feature fusion (VIs+TFs+TIs) effectively estimated LAI, achieving an R2 of 0.859 and 0.794”
t0“Estimating Maize Leaf Area Index Using Multi-Source Features Derived from UAV Multispectral Imagery and Machine Learning Models”