Application of Multitemporal Change Detection in Radar Satellite Imagery Using REACTIV-Based Method for Geospatial Intelligence

Sensors (Basel). 2023 May 19;23(10):4922. doi: 10.3390/s23104922.

Abstract

Constant monitoring of airports and aviation bases has become one of the priorities in today's strategic security. It results in the necessity to develop the potential of satellite Earth observation systems and to intensify the efforts to develop the technologies of processing SAR data, in particular in the aspect of detecting changes. The aim of this work is to develop a new algorithm based on the modified core REACTIV in the multitemporal detection of changes in radar satellite imagery. For the purposes of the research works, the new algorithm implemented in the Google Earth Engine environment has been transformed so that it would meet the requirements posed by imagery intelligence. The assessment of the potential of the developed methodology was performed based on the analysis of the three main aspects of change detection: analysis of infrastructural changes, analysis of military activity, and impact effect evaluation. The proposed methodology enables automated detection of changes in multitemporal series of radar imagery. Apart from merely detecting the changes, the method also allows for the expansion of the change analysis result by adding another dimension: the determination of the time of the change.

Keywords: REACTIV; SAR; change detection; coefficient of variation; geospatial intelligence; multitemporal analysis; time series.

Grants and funding

This work was financed by the Military University of Technology, Faculty of Civil Engineering and Geodesy, under research project UGB grant number 815/2023.