Satellite Images Analysis with Symbolic Time Series: A Case Study of the Algerian Zone

TL;DR

This paper explores the use of symbolic time series representation, specifically SAX, to analyze satellite image time series for land use change detection over five years.

Contribution

It applies the SAX symbolic representation to satellite image time series, demonstrating its effectiveness in reducing dimensionality and capturing land evolution.

Findings

SAX effectively summarizes satellite time series data.

Symbolic representation aids in land change analysis.

Dimensionality reduction improves data processing efficiency.

Abstract

Satellite Image Time Series (SITS) are an important source of information for studying land occupation and its evolution. Indeed, the very large volumes of digital data stored, usually are not ready to a direct analysis. In order to both reduce the dimensionality and information extraction, time series data mining generally gives rise to change of time series representation. In an objective of information intelligibility extracted from the representation change, we may use symbolic representations of time series. Many high level representations of time series have been proposed for data mining, including Fourier transforms, wavelets, piecewise polynomial models, etc. Many researchers have also considered symbolic representations of time series, noting that such representations would potentiality allow researchers to avail of the wealth of data structures and algorithms from the text processing and bioinformatics communities. We present in this work, one of the main symbolic representation methods "SAX"(Symbolic Aggregate Approximation) and we experience this method to symbolize and reduce the dimensionality of a Satellite Image Times Series acquired over a period of 5 years by characterizing the evolution of a vegetation index (NDVI).