Machine Learning-Based Approaches for Predicting SPAD Values of Maize Using Multi-Spectral Images

Precisely monitoring the growth condition and nutritional status of maize is crucial for optimizing agronomic management improving agricultural production. Multi-spectral sensors are widely applied in ecological domains. However, images collected under varying weather conditions on multiple days show a lack data consistency. In this study, Mini MCA 6 Camera from UAV platform was used to collect covering different stages maize. The empirical line calibration method establish generic equations radiometric calibration. coefficient determination (R2) reflectance calibrated ASD Handheld-2 ranged 0.964 0.988 (calibration), 0.874 0.927 (validation), respectively. Similarly, root mean square errors (RMSE) were 0.110, 0.089, 0.102% validation using 5 August, 21 September, both 2019, soil plant analyzer development (SPAD) values measured build linear regression relationships with spectral textural indices stages. Stepwise model (SRM) identify optimal combination estimating SPAD values. support vector machine (SVM) random forest (RF) models independently based combinations. SVM performed better than RF R2 (0.81) RMSE (0.14), This study contributed retrieval extracted multi-spectral learning methods.