Monitoring the maturity of multi-species maize based on remote sensing and thus mastering the optimal harvesting time is crucial for improving its yield and quality. The traditional method to monitor the maturity progress of maize is to use field surveys, and the disappearance of the kernel "milkline" is usually taken as a sign of maturity. However, the traditional field survey method is a labor-intensive activity that is not conducive to high-throughput field monitoring. Therefore, this study aims to construct a maize maturity index (MMI) to quantify the maturity of maize and monitor it through UAV multispectral monitoring, so as to grasp the dynamics of maize maturity stage in the field. Firstly, the UAV platform was used to acquire multispectral images at five time points of the maize maturity stage, and ground-based measured data such as the percentage of milkline, kernel water content and leaf chlorophyll content were collected accordingly. Secondly, based on the weighted analysis of the measured data, the MMI was constructed. Finally, based on the MMI and the vegetation index, a model was constructed using regression models and random forests to realize the UAV multispectral monitoring of corn maturity, and the effects of different varieties on MMI were analyzed. The results showed that: 1) for different varieties of maize at maturity stage, there were differences in the change patterns of leaf chlorophyll content and kernel water content, the leaf chlorophyll content and kernel water content of Zhengdan 958 and Jingjiuqingzhu16 were always higher than that of Jiyuan 1 and Jiyuan 168, while the rate of decline of leaf chlorophyll content and milkline percentage of two varieties of maize at maturity stage was lower than that of Jiyuan 1 and Jiyuan 168. 2) The correlations between MMI and selected vegetation indices in the experiment could reach 0.01 significant level, among which the correlations with normalized difference vegetation index (NDVI) and transformed chlorophyll absorbtion ratio index (TCARI) were highest with correlation coefficients above 0.87, in addition, the wide dynamic range vegetation index (WDRVI) has the most obvious changes, and the variance fluctuates less, which is similar to MMI. 3) The study was verified based on data sets of different combinations. Among them, the random forest model has the highest estimation accuracy of MMI. The test set coefficient of determination (R²) is 0.84, and the root mean squared error (RMSE) is 8.77%, and the normalized root mean squared error (nRMSE) is 12.05%. In the revalidation scheme 2-1, the RF model test set has high accuracy, in which R² is 0.65, RMSE is 13.02%, nRMSE is 19.17%. In addition, the random forest model has better estimated accuracy of different varieties of MMI. The Jingjiuqingzhu 16 has the best accuracy. Among them, R², RMSE, and nRMSE are 0.76, 10.67%, and 15.88%. The model accuracy proves that the drone platform can be monitored to monitor the maturity of different varieties of corn. The results of the research can provide a reference for the dynamic changes in multi-spectrum drones to monitor the dynamic changes in multi-variety of corn in farmland.

Accurate monitoring and spatial distribution of the leaf chlorophyll content (LCC) and canopy chlorophyll content (CCC) of individual apple trees are highly important for the effective management of individual plants and the promotion of the construction of modern smart orchards. However, the estimation of LCC and CCC is affected by shadows caused by canopy structure and observation geometry. In this study, we resolved the response relationship between individual apple tree crown spectra and shadows through a three-dimensional radiative transfer model (3D RTM) and unmanned aerial vehicle (UAV) multispectral images, assessed the resistance of a series of vegetation indices (VIs) to shadows and developed a hybrid inversion model that is resistant to shadow interference. The results revealed that (1) the proportion of individual tree canopy shadows exhibited a parabolic trend with time, with a minimum occurring at noon. Correspondingly, the reflectance in the visible band decreased with increasing canopy shadow ratio and reached a maximum value at noon, whereas the pattern of change in the reflectance in the near-infrared band was opposite that in the visible band. (2) The accuracy of chlorophyll content estimation varies among different VIs at different canopy shadow ratios. The top five VIs that are most resistant to changes in canopy shadow ratios are the NDVI-RE, Cire, Cigreen, TVI, and GNDVI. (3) For the constructed 3D RTM ​+ ​GPR hybrid inversion model, only four VIs, namely, NDVI-RE, Cire, Cigreen, and TVI, need to be input to achieve the best inversion accuracy. (4) Both the LCC and the CCC of individual trees had good validation accuracy (LCC: R² ​= ​0.775, RMSE ​= ​6.86 ​μg/cm², nRMSE ​= ​12.24 ​%; CCC: R² ​= ​0.784, RMSE ​= ​32.33 ​μg/cm², and nRMSE ​= ​14.49 ​%), and their distributions at orchard scales were characterized by considerable spatial heterogeneity. This study provides ideas for investigating the response between individual tree canopy shadows and spectra and offers a new strategy for minimizing the influence of shadow effects on the accurate estimation of chlorophyll content in individual apple trees.

Accurate and real-time monitoring true leaf area index (LAI) is an essential for assessing crop growth status and predicting yields. Conventional LAI inversion approaches have been constrained by insufficient data representativeness and environmental variability, particularly when applied across interannual variations and different phenological stages. This study presented a novel methodology integrating three-dimensional radiative transfer modeling (3D RTM) with knowledge-guided deep learning to address these limitations. We developed a knowledge-guided convolutional neural network (KGCNN) architecture incorporating 3D canopy structural physics, enhanced through transfer learning (TL) techniques for cross-temporal adaptation. The KGCNN model was initially pre-trained on synthetic datasets generated by the large-scale remote sensing scattering model (LESS), followed by domain-specific fine-tuning using 2021 field measurements, and culminating in cross-year validation with 2022-2023 datasets. Our results demonstrated significant improvements over conventional approaches, with the 3D RTM-based KGCNN achieving superior performance compared to 1D RTM implementations (PROSAIL + CNN + TL). Specially, for the 2022 dataset, the overall R² increased by 0.27 and RMSE decreased by 2.46; for the 2023 dataset, the overall RMSE decreased by 1.62, compared to the PROSAIL ​+ ​TL method. Our method (3D RTM ​+ ​KGCNN ​+ ​TL) delivered superior LAI retrieval accuracy on the two-year datasets compared to LSTM ​+ ​TL, RNN ​+ ​TL, and 3D RTM ​+ ​RF models. This study also introduced an effective 3D scene modeling strategy that integrates scenarios representing the measured data range with additional synthetic scenes generated through random combinations of structural parameters. By incorporating detailed 3D crop structural information into the KGCNN network and fine-tuning the model with measured data, the approach significantly enhanced the model's adaptability to varying data distributions across different years and growth stages. This approach thus improved both the accuracy and stability of true LAI retrieval.