Satellite Image Time Series Decomposition Based on EEMD

Yun-Long Kong,Yuan Yuan,Wei Li,An-Zhi Yue,Yu Meng

doi:10.3390/rs71115583

Abstract

Satellite Image Time Series (SITS) have recently been of great interest due to the emerging remote sensing capabilities for Earth observation. Trend and seasonal components are two crucial elements of SITS. In this paper, a novel framework of SITS decomposition based on Ensemble Empirical Mode Decomposition (EEMD) is proposed. EEMD is achieved by sifting an ensemble of adaptive orthogonal components called Intrinsic Mode Functions (IMFs). EEMD is noise-assisted and overcomes the drawback of mode mixing in conventional Empirical Mode Decomposition (EMD). Inspired by these advantages, the aim of this work is to employ EEMD to decompose SITS into IMFs and to choose relevant IMFs for the separation of seasonal and trend components. In a series of simulations, IMFs extracted by EEMD achieved a clear representation with physical meaning. The experimental results of 16-day compositions of Moderate Resolution Imaging Spectroradiometer (MODIS), Normalized Difference Vegetation Index (NDVI), and Global Environment Monitoring Index (GEMI) time series with disturbance illustrated the effectiveness and stability of the proposed approach to monitoring tasks, such as applications for the detection of abrupt changes.

Highlights

With the development of satellite technology, the amount of historical remote sensing data is growing
We presented a decomposition approach for Satellite Image Time Series (SITS) based on Ensemble Empirical Mode Decomposition (EEMD)
The proposed method first decomposed time series into several Intrinsic Mode Functions (IMFs) and a residue, and chose appropriate IMFs to make up the seasonal component according to the spectral characteristics of the targets and the trend component according to the time scale

Summary

Introduction

With the development of satellite technology, the amount of historical remote sensing data is growing. These long sequences of remote sensing data, which are called Satellite Image Time Series (SITS), contain a considerable amount of information, such as land-cover dynamics and disturbances. SITS usually contain seasonal, trend, and remainder components [2]. Seasonal and trend components are key elements for SITS analysis because they contain important information for understanding Earth science processes. Seasonal components constitutionally reflect inter-annual periodic fluctuations that are usually driven by the annual temperature, weather, seasons of the year, or growth cycles of vegetation [3]. Decomposing SITS into different components, especially the seasonal components and trend components, plays an important role in SITS analysis. Various techniques were investigated for this purpose

Objectives

Discussion

Conclusion