Timely and accurate in-season estimation of aboveground biomass (AGB) and yield in winter wheat is crucial for optimizing resources and ensuring food security. Light use efficiency (LUE) models have proven effective in estimating crop gross primary productivity and yield across sites and years due to their strong physiological and ecological mechanisms. However, existing studies are limited to satellite applications and have not utilized unmanned aerial vehicle (UAV) imagery. This study proposed a practical framework for accurate in-season estimation of AGB and yield in winter wheat from UAV imagery by combining a LUE model and machine learning (LUE-ML) across five plot experiments. Subsequently, the scalability of the LUE-ML yield prediction approach was assessed in farmer's fields from five counties of Jiangsu Province, China. The results demonstrated that while the AGB for the heading stage was estimated by combining the retrieved LAI and 20-day accumulated meteorological features, the AGB during the post-heading period could be estimated accurately using the stage-skipping or stage-progressive strategy, with the latter (R_val² = 0.93) outperforming the former (R_val² = 0.84). The combination of one spectral index, LUE-derived AGB, and three 20-day accumulated relative meteorological features (Comb. #6) performed the best (R_cal² = 0.89; R_val² ≥ 0.79) for yield prediction among all combinations. When extended to farmer-field yield prediction across the province, Comb. #6 also achieved acceptable performance. This study suggests the use of LUE-ML models represents a significant step forward towards mechanistic estimation of AGB and yield for cereal crops from UAV imagery.

Rapid and accurate estimation of panicle number per unit ground area (PNPA) in winter wheat before heading is crucial to evaluate yield potential and regulate crop growth for increasing the final yield. The accuracies of existing methods were low for estimating PNPA with remotely sensed data acquired before heading since the spectral saturation and background effects were ignored. This study proposed a spectral-textural PNPA sensitive index (SPSI) from unmanned aerial vehicle (UAV) multispectral imagery for reducing the spectral saturation and improving PNPA estimation in winter wheat before heading. The effect of background materials on PNPA estimated by textural indices (TIs) was examined, and the composite index SPSI was constructed by integrating the optimal spectral index (SI) and TI. Subsequently, the performance of SPSI was evaluated in comparison with other indices (SI and TIs). The results demonstrated that green-pixel TIs yielded better performances than all-pixel TIs apart from TI_[HOM], TI_[ENT], and TI_[SEM] among all indices from 8 types of textural features. SPSI, which was calculated by the formula DATT_{[850,730,675]} + NDTI_COR[850,730], exhibited the highest overall accuracies for any date in any dataset in comparison with DATT_{[850,730,675]}, TI_NDRE[MEA], and NDTI_COR[850,730]. For the unified models assembling 2 experimental datasets, the R_V² values of SPSI increased by 0.11 to 0.23, and both RMSE and RRMSE decreased by 16.43% to 38.79% as compared to the suboptimal index on each date. These findings indicated that the SPSI is valuable in reducing the spectral saturation and has great potential to better estimate PNPA using high-resolution satellite imagery.