deFOREST: Fusing Optical and Radar satellite data for Enhanced Sensing of Tree-loss

2025-10-15

AI Summary

This study addresses the challenge of cloud- and rain-induced degradation in optical remote sensing data, which severely compromises forest loss monitoring. To overcome this limitation, we propose a temporally aware anomaly detection framework that deeply fuses optical and SAR data. Methodologically, we construct a prior-free optical residual space via discrete Karhunen–Loève expansion to generate robust anomaly maps; subsequently, we integrate Sentinel-2 optical and Sentinel-1 SAR time series, leveraging concentration inequalities and a hidden Markov model (HMM) for pixel-wise deforestation classification. Our key innovation lies in the first coupling of distribution-free residual modeling with SAR-assisted correction, markedly enhancing detection reliability under heavy cloud cover. Evaluated across large-scale Amazonian regions, the method surpasses current state-of-the-art approaches in accuracy and demonstrates exceptional robustness even under sparse optical data conditions.

Technology Category

• Computer Vision: Remote Sensing / Geospatial AI
• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Machine Learning: Calibration & Uncertainty Quantification

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs
• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

In this paper we develop a deforestation detection pipeline that incorporates optical and Synthetic Aperture Radar (SAR) data. A crucial component of the pipeline is the construction of anomaly maps of the optical data, which is done using the residual space of a discrete Karhunen-Loève (KL) expansion. Anomalies are quantified using a concentration bound on the distribution of the residual components for the nominal state of the forest. This bound does not require prior knowledge on the distribution of the data. This is in contrast to statistical parametric methods that assume knowledge of the data distribution, an impractical assumption that is especially infeasible for high dimensional data such as ours. Once the optical anomaly maps are computed they are combined with SAR data, and the state of the forest is classified by using a Hidden Markov Model (HMM). We test our approach with Sentinel-1 (SAR) and Sentinel-2 (Optical) data on a $92.19,km imes 91.80,km$ region in the Amazon forest. The results show that both the hybrid optical-radar and optical only methods achieve high accuracy that is superior to the recent state-of-the-art hybrid method. Moreover, the hybrid method is significantly more robust in the case of sparse optical data that are common in highly cloudy regions.

Problem

• Fusing optical and radar satellite data for deforestation detection
• Developing anomaly maps without requiring prior data distribution knowledge
• Enhancing robustness in deforestation monitoring under cloudy conditions

Innovation

• Fusing optical and radar satellite data for deforestation detection
• Using KL expansion residuals for anomaly mapping without distribution assumptions
• Combining anomaly maps with SAR data via Hidden Markov Model