Bidirectional recurrent imputation and abundance estimation of LULC classes with MODIS multispectral time series and geo-topographic and climatic data

TL;DR

This paper introduces a novel deep learning approach using LSTM models with geo-topographic and climatic data to improve land cover abundance estimation from MODIS multispectral time series, eliminating the need for explicit endmember extraction.

Contribution

It pioneers spectral unmixing with MODIS multispectral time series using end-to-end deep learning that incorporates ancillary data, and provides a new labeled dataset for the region of Andalusia.

Findings

Integrating spectral-temporal data with geo-topographic and climatic information improves abundance estimation.

The proposed model outperforms traditional methods in accuracy.

The Andalusia-MSMTU dataset is publicly available for further research.

Abstract

Remotely sensed data are dominated by mixed Land Use and Land Cover (LULC) types. Spectral unmixing (SU) is a key technique that disentangles mixed pixels into constituent LULC types and their abundance fractions. While existing studies on Deep Learning (DL) for SU typically focus on single time-step hyperspectral (HS) or multispectral (MS) data, our work pioneers SU using MODIS MS time series, addressing missing data with end-to-end DL models. Our approach enhances a Long-Short Term Memory (LSTM)-based model by incorporating geographic, topographic (geo-topographic), and climatic ancillary information. Notably, our method eliminates the need for explicit endmember extraction, instead learning the input-output relationship between mixed spectra and LULC abundances through supervised learning. Experimental results demonstrate that integrating spectral-temporal input data with geo-topographic and climatic information significantly improves the estimation of LULC abundances in mixed pixels. To facilitate this study, we curated a novel labeled dataset for Andalusia (Spain) with monthly MODIS multispectral time series at 460m resolution for 2013. Named Andalusia MultiSpectral MultiTemporal Unmixing (Andalusia-MSMTU), this dataset provides pixel-level annotations of LULC abundances along with ancillary information. The dataset (https://zenodo.org/records/7752348) and code (https://github.com/jrodriguezortega/MSMTU) are available to the public.