Field experiments were conducted to analyze the effectiveness of the crop stress index (CWSI) obtained by infrared thermal imaging to indicate crop water status, and to determine the appropriate CWSI threshold range for wheat at different growth stages. The results showed that the sensitivity of plant physiological parameters to soil water was different at different growth stages. The sensitivity of stomatal conductance (Gs) and transpiration rate (Tr) to soil water was higher than that of leaf relative water content (LRWC) and photosynthetic rate (Pn). The characteristics of plant physiology and biomass (yield) at each growth stage showed that the plant production would not suffer from drought stress as long as the soil water content (SWC) was maintained above 57.0% of the field water capacity (FWC) during the jointing stage, 63.0% of the FWC during the flowering stage and 60.0% of the FWC during the filling stage. Correlation analysis showed that the correlation of CWSI with Gs, Tr and Pn was lower than that with LRWC and SWC at the jointing stage. CWSI was extremely significantly negatively correlated with SWC and LRWC (p < 0.01), but significantly negatively correlated with Gs, Tr and Pn (p < 0.05). At the flowering stage, CWSI was extremely significantly negatively correlated with all physiological and soil parameters (p < 0.01). The regression analysis showed that the CWSI of winter wheat was correlated with biomass (grain yield) in a curvilinear relationship at each growth stage. When the CWSI increased to a certain extent, the biomass and yield showed a decreasing trend with the increase in CWSI. Comprehensive analysis of all indexes showed that CWSI can be used as a decision-making index to guide the water-saving irrigation of winter wheat, as long as the CWSI threshold of plants was maintained at 0.26–0.38 during the jointing stage, 0.27–0.32 during the flowering stage and 0.30–0.36 during the filling stage, which could not only avoid the adverse effects of water stress on crop production, but also achieve the purpose of water saving.

(1) Background: Accurate inversion of crop water content is key to making an intelligent irrigation decision. However, little effort has been devoted to accurately estimating the crop water content of winter wheat in the North China Plain. (2) Method: The crop water content of winter wheat was measured at jointing, flowering and grain-filling stages, respectively. UAV-based multispectral remote sensing images were used to calculate thirteen vegetation indices, including SAVI, EVI, R-M, NDRE, OSAVI, GOSAVI, REOSAVI, GBNDVI, NDVI, RVI, DVI, GNDVI, and TVI. Five machine learning (ML) algorithms (i.e., MLR, RF, PLSR, ElasticNet, and ridge regression) were adopted to estimate the crop water content of winter wheat at the three growth stages. The benchmark datasets, which include CWC as well as vegetation indices calculated based on spectral indices, were adopted to validate the performance of the ML models. (3) Results: The correlation coefficients ranged from 0.64 to 0.82 at different growth stages. The optimal vegetation indices were GNDVI for the jointing stage, NDRE for the flowering and the grain-filling stage, respectively. Among the five machine learning methods, random forest (RF) showed the best performance across the three growth stages, with its coefficient of determination (R²) of 0.80, or an increase by 20.1% than those of other models. In addition, the RMSE and RPD of the RF model at the flowering stage were 3.00% and 2.01, which significantly outperformed other models and growth stages. (4) Conclusion: This study may provide theoretical support and technical guidance for monitoring current water status in wheat crops, which is useful to develop a precise irrigation prescription map for local farmers. (5) Limitation: The main limitation of this study is that the sample size is relatively small and may not fully reflect the characteristics of the target groups. At the same time, subjectivity and bias may exist in the data collection, which may have a certain impact on the accuracy of the results. Future studies could consider expanding sample sizes and improving data collection methods to overcome these limitations.