Estimation of Summer Maize Growth Parameters Based on Multi-Source Data Fusion of Ground-Air Integration

Chlorophyll content and leaf area index (LAI), as important parameters reflecting the change in maize growth, are important indicators for monitoring maize growth status and yield prediction. At present, low accuracy is still a problem in the growth monitoring and yield prediction of summer maize in the field. This paper proposes a method to accurately monitor LAI and the soil plant analysis development (SPAD) values (Which strongly correlate with leaf chlorophyll contents) based on ground-air multi-source data fusion. It could provide a scientific basis for large-area, rapid and efficient monitoring of maize growth. In 2020 and 2021, we had been collected the Unmanned Aerial Vehicle (UAV) multispectral data, ground hyperspectral data respectively, UAV visible light data and environmental accumulated temperature data in multiple growth periods of summer maize. The effective plant height, effective accumulation temperature and canopy vegetation index of maize were calculated. And then, we extracted the hyperspectral features of the maize canopy and optimized the LAI and SPAD sensitive features through correlation analysis. Based on the single-source and multi-source data, the multiple linear regression (MLR), the partial least squares regression (PLSR) and random forest (RF) regression were used to construct an inversion model of LAI and SPAD. After that, the best model was used to generate the LAI and SPAD distribution prescription map. Finally, we examined the changing trend of the LAI and SPAD. The results showed that the correlations between the position of the hyperspectral red edge and the first-order differential value in the red edge with LAI and SPAD were all greater than 0.5. The correlation between vegetation index including a red and near-infrared band with LAI and SPAD was above 0.75. The correlation between crop height and effective accumulated temperature with LAI and SPAD was above 0.7. The inversion models based on multi-source data were more effective than the models with single data. The RF model with multi-source data fusion achieved the highest accuracy of all models. In the testing set, the LAI and SPAD model R² were 0.9315 and 0.7767, the RMSE were 0.4895 and 2.8387. The absolute error between the extraction result of each model prescription map and the measured value was small. The error between the predicted value and the measured value of the LAI prescription map generated by the RF model was less than 0.4895. The difference between the predicted value and the measured value of the SPAD prescription map was less than 2.8387. The LAI and SPAD of summer maize increased first and then decreased with the advancement of the growth period, which was in line with the actual growth conditions. The research results indicate that the proposed method could effectively monitor maize growth parameters and provide a scientific basis for summer maize field management.