A Method to Automatically Detect Changes in Multitemporal Spectral Indices: Application to Natural Disaster Damage Assessment

Luca Pulvirenti,Giuseppe Squicciarino,Elisabetta Fiori

doi:10.3390/rs12172681

Luca Pulvirenti, Giuseppe Squicciarino + Show 1 more

Open Access

https://doi.org/10.3390/rs12172681

Copy DOI

Journal: Remote sensing	Publication Date: Aug 19, 2020
Citations: 8	License type: CC BY 4.0

Affiliation: CIMA Research Foundation

Abstract

This paper presents a new method, based on clustering and thresholding, to automatically perform binary change detection in multitemporal spectral indices. The method is denoted as Buffer-From-Cluster Approach (BFCA). To estimate the distributions of changed and unchanged pixels, as needed for the purpose of a reliable thresholding of a spectral index, a clustering algorithm is preliminarily applied to identify image objects possibly corresponding to areas where significant changes occurred. Then, a buffer zone is created around the selected cluster to identify unchanged areas surrounding changed ones. The cluster and the buffer zone are jointly analyzed to estimate the distributions of changed and unchanged pixels and to verify that they can be distinguished from each other. Finally, the results of thresholding and clustering are combined to generate the binary change map. The BFCA has been conceived to map the extent of the areas affected by a natural disaster like wildfire. To validate the proposed method, burned area maps produced by applying the BFCA to spectral indices derived from Sentinel-2 data have been compared to maps produced by the Copernicus Emergency Management Service. For testing the multi-hazard detection capability, the same kind of exercise has been carried out for a flooding test case too. The positive results of the comparison have confirmed the effectiveness of the proposed method.

Highlights

The frequency and the destructivity of natural hazards is continuously increasing
This paper describes the various phases of the Buffer-From-Cluster Approach (BFCA) and presents the results of its application to S2-derived multitemporal spectral index (MTSI) useful to map burned area (BA) and floods
It includes the intermediate results of the various steps of the BFCA and the final BA map

Summary

Introduction

The frequency and the destructivity of natural hazards is continuously increasing They leave in their wake a trail of injury, death, loss of livestock, property damage, and economic loss. This implies the need for improving the capability to manage these events. Operational services like the rapid mapping component of the Copernicus Emergency Management Service (CEMS) currently provide emergency managers with maps of natural disasters at different spatial resolutions depending on the kind of EO data used to produce the maps. The present availability of Sentinel-1/2/3 data allows for routinely producing maps of areas affected by disasters (without the need of waiting for an activation). Fully automatic algorithm able to rapidly map the disaster’s extent is needed to accomplish this routine production, especially if many EO data must be sequentially processed (e.g., when working at large spatial scales)

Methods

Results

Conclusion