Multilayer Architecture for Heterogeneous Geospatial Data Analytics: Querying and Understanding EO Archives

Abstract

The constantly growing process of the Earth Observation (EO) data and their heterogeneity require new systems and tools for effectively querying and understanding the available data archives. In this paper we present a tool for heterogeneous geospatial data analytics. The system implements different web technologies in a multilayer server-client architecture allowing the user to visually analyze satellite images, maps and in-situ information. Specifically, the information managed is composed of EO multispectral and SAR products along with the multitemporal in-situ LUCAS surveys. The integration of these data provides a very useful information during the EO scene interpretation process. The system also offers interactive tools for the detection of optimal datasets for EO multitemporal image change detection, providing at the same time ground truth points for both, human and machine analysis. Furthermore, we show by means of visual analytic representations a way to analyze and understand the content and distribution of the EO databases.

Mining Multitemporal In Situ Heterogeneous Monitoring Information for the Assurance of Recorded Land Cover Changes

“…We have presented a data mining methodology that is able to successfully filter false land cover changes from the real land cover change detections in multitemporal LUCAS in-situ surveys. We shortly introduced the heterogeneous geospatial data analytics system [7], which is the base platform of the presented data mining methodology. We have described the three methodology steps and evaluated them independently.…”

“…The base of the presented mining methodology is the system for heterogeneous geospatial data analytics first presented in [7]. The system, as shown in Fig.…”

“…The client is constituted of the Graphic User Interface (GUI) layer accessible from any electronic device capable of running a HTML-5 compatible web browser. A detailed description of the architecture and its functionalities can be found in [7].…”

“…Regarding the use of in-situ data, the work in [6] integrates distributed in-situ data with very high resolution optical EO images providing: geospatial data queries, on demand image processing, and fast map visualizations to support collaborative and more efficient emergency response. Finally, in [7] a web based GIS for visual land cover analytics was presented. The system architecture integrated in-situ data from the European Land Use and Coverage Area frame Survey (LUCAS) [8] with different EO products, e.g., Synthetic Aperture Radar (SAR) images for German satellite TerraSAR-X.…”

Review on Integrating Geospatial Big Datasets and Open Research Issues

“…Nevertheless, further research is needed to determine which geospatial applications are most influential as well as to integrate geospatial techniques and parallelization in this age of big data [100]. The lack of big data homogeneity poses a particular research difficulty to formulate more advanced algorithms to be used more effectively [101].…”