A Condition-Aware Shading Domain-Adaptive Framework for Robust Chlorophyll Inversion Across Shade Managements in...

Area of Science:

• Plant Science
• Remote Sensing
• Agricultural Technology

Background:

• Shade management in cultivation and regeneration alters plant light environments.
• This significantly challenges accurate plant phenotyping, particularly chlorophyll estimation.
• Trait inversion performance is degraded under variable light conditions.

Purpose of the Study:

• To develop a robust method for chlorophyll retrieval in Hopea hainanensis under shade management.
• To address the challenge of illumination-regime-dependent domain shift in plant phenotyping.
• To improve the accuracy and stability of chlorophyll content estimation in heterogeneous light environments.

Main Methods:

• Reframing chlorophyll retrieval as a conditional domain shift problem.
• Developing a condition-aware domain adaptation framework (CAI-DAI).
• Integrating conditional encoding and alignment for improved model performance.

Main Results:

• Chlorophyll content increased with shading intensity, showing condition-dependent spectral variations.
• CAI-DAI and CA-IE outperformed comparative models across various fine-tuning ratios.
• CAI-DAI demonstrated the best and most stable performance, with low error metrics (MAE, nRMSE) and good R².
• CAI-DAI showed narrower error ranges and smaller fluctuations compared to CA-IE across individual shading levels.

Conclusions:

• The proposed CAI-DAI framework effectively enhances robustness for chlorophyll monitoring under heterogeneous shading.
• The method provides methodological support for shade management decisions with limited labeled data.
• Accurate chlorophyll estimation is crucial for understanding plant responses to varying light environments.