Super-resolution-enhanced texture and vegetation index fusion for drone-based detection of corn growth status under waterlogging stress

Waterlogging stress significantly affects crop physiology, particularly chlorophyll content, which is a key indicator of photosynthetic capacity and crop health. However, existing studies often rely primarily on spectral information and fail to fully capture the structural-physiological relationship between canopy texture and chlorophyll content under stress conditions. This study proposes a super-resolution-enhanced feature fusion framework that integrates vegetation indices and texture features derived from UAV multispectral imagery. Super-resolution techniques were applied to improve spatial detail and enhance the sensitivity of texture descriptors to canopy structural variations. The fused features were then used to estimate chlorophyll content under different waterlogging conditions. Results show that combining texture and vegetation indices slightly improves prediction performance compared with single-feature models. Super-resolution further enhances the correlation between texture features and chlorophyll content, leading to more consistent model performance. Among the tested approaches, the proposed framework achieved stable prediction accuracy (RC2 = 0.74, RV2 = 0.74, RMSEC = 3.36 mg L−1, RMSEV = 3.95 mg L−1), indicating its effectiveness for chlorophyll estimation. This study provides new insight into how super-resolution enhances the physiological interpretability of texture features in UAV-based crop monitoring, offering a promising approach for non-destructive assessment of crop growth under waterlogging stress.