A Scalable Cloud Computing Infrastructure for Geospatial Data Analytics for Change Detection

Geospatial data analytics is emerging as a promising technology in decision support systems in many application domains. This paper presents results from a study on the use of elastic cloud computing to provide geospatial data analytics in an efficient way. The analytics is applied to bulks of satellite data from multi-spectral optical sensors. The system design includes data integration, data caching, and a job manager to orchestrate the data analytics workflow. The system is validated in a case study focusing on precision agriculture, where a simple harvest detection algorithm is implemented to identify the time of harvest for crop fields. The performance of the underlying cloud computing infrastructure is evaluated and the benefit of scalingthe cloud horizontally when bulk satellite data becomes available is discussed.