Spectral temporal unmixing improves MODIS data analysis accuracy
Multi-task Disentangling Fusion Mamba-CNN for MODIS Spectral-Temporal Unmixing with Endmember Variability Modeling
Computer Vision and Pattern Recognition
Summary
Analyzing satellite data from MODIS sensors is tricky because the color information changes over time and different materials can look very similar. The authors created a new method called MDF-Mamba that separates these time and color effects using expert modules for each color channel, combined with a special branch that captures seasonal changes. This method also learns how materials vary within the same type, improving the accuracy of identifying the components in each pixel. Their tests over large geographic areas showed better results than previous methods.
What this means in practice
- •For environmental monitoring teams: Map land cover abundances more accurately by separating mixed satellite signals over time to monitor environmental changes in regions like Alberta and Saskatchewan.
- •For remote sensing software developers: Incorporate a stable fusion approach to combine spectral and temporal data in their software for improved spectral library creation from MODIS data.
Authors
Motasem Alkayid, Yimin Zhu, Naser El-Sheimy, Lincoln Linlin Xu
Abstract
Spectral-temporal unmixing of MODIS time series is essential for applications such as abundance mapping and spectral library construction, yet it remains challenging due to (i) spectral-temporal information coupling in MODIS data, (ii) intra-class endmember spectral variability and inter-class signature ambiguity, and (iii) the instability of naively fusing multiple expert representations. This paper presents the Multi- task Disentangling Fusion Mamba (MDF-Mamba) to address these limitations. To resolve spectral-temporal coupling, we design a disentangling architecture that pairs per-channel spectral experts with a dedicated bidirectional state-space (Mamba) temporal branch. Convolutional modules each process a single spectral channel, while the Mamba branch encodes the cross-channel seasonal trajectory, forming a multi-branch expert ensemble with soft inter-expert reweighting based on squeeze-and-excitation. For the integration of these spectral and temporal expert outputs, we introduce mHCFusion, adapting the manifold-constrained hyper-connection framework to merge the outputs in a stable and interpretable manner. Endmember variability is addressed by extending the linear mixing model with pixel-wise learnable spectral variability weights that capture intra-class endmember variation. Finally, we design a multi-task training objective that combines supervised abundance estimation with self-supervised spectral reconstruction and endmember regularization terms on the same training data, yielding a highly efficient framework for spectral-temporal unmixing. Experiments on MODIS time series covering the Alberta and Saskatchewan provinces show that MDF-Mamba achieves the best performance on 12 of 14 evaluation metrics across both datasets (7 of 7 on Alberta; 5 of 7 on Saskatchewan), ranking second on the remaining two.