Cloud Optical Thickness Retrievals Using Angle Invariant Attention Based Deep Learning Models

2025-05-30

AI Summary

To address the limited accuracy of cloud optical thickness (COT) retrieval in satellite remote sensing—caused by 3D cloud effects, viewing geometry variations, and atmospheric interference—this paper proposes an angle-invariant attention-based deep learning model. The method innovatively integrates angular encoding with self-attention mechanisms to explicitly model satellite observation geometry and three-dimensional radiative transfer processes. A multi-angle collaborative training strategy is further introduced to enhance robustness against variations in solar and viewing zenith angles. Trained end-to-end on multi-view radiative transfer simulations, the proposed model reduces COT retrieval error by at least a factor of nine compared to state-of-the-art deep learning approaches under complex cloud conditions and large zenith angles. This yields significantly improved generalization capability and stronger physical consistency with underlying radiative transfer principles.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Calibration & Uncertainty Quantification
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Cloud Optical Thickness (COT) is a critical cloud property influencing Earth's climate, weather, and radiation budget. Satellite radiance measurements enable global COT retrieval, but challenges like 3D cloud effects, viewing angles, and atmospheric interference must be addressed to ensure accurate estimation. Traditionally, the Independent Pixel Approximation (IPA) method, which treats individual pixels independently, has been used for COT estimation. However, IPA introduces significant bias due to its simplified assumptions. Recently, deep learning-based models have shown improved performance over IPA but lack robustness, as they are sensitive to variations in radiance intensity, distortions, and cloud shadows. These models also introduce substantial errors in COT estimation under different solar and viewing zenith angles. To address these challenges, we propose a novel angle-invariant, attention-based deep model called Cloud-Attention-Net with Angle Coding (CAAC). Our model leverages attention mechanisms and angle embeddings to account for satellite viewing geometry and 3D radiative transfer effects, enabling more accurate retrieval of COT. Additionally, our multi-angle training strategy ensures angle invariance. Through comprehensive experiments, we demonstrate that CAAC significantly outperforms existing state-of-the-art deep learning models, reducing cloud property retrieval errors by at least a factor of nine.

Problem

• Improving Cloud Optical Thickness (COT) retrieval accuracy
• Addressing biases from 3D cloud effects and viewing angles
• Enhancing robustness of deep learning models for COT estimation

Innovation

• Angle-invariant attention-based deep learning model
• Angle embeddings for satellite viewing geometry
• Multi-angle training strategy for robustness