Conference Paper

Cloud Architecture for Processing and Distribution of Satellites Imagery

Abstract: 

Earth Observation (EO) records information of the World to be used in many applications: emergency services, precision agriculture, borders control, environmental monitoring, infrastructures control and many others. The base of that information is satellite imagery. This provides images of the world that are analyzed and distributed to the end users.
New EO services require higher amount of data, more flexibility and instant access to the products and services. This is difficult to achieve with traditional on premises infrastructures, but cloud computing can provide its flexibility, scalability and on demand use characteristics to the EO field.
We propose a novel architecture of a satellite imagery data center specifically designed for its implementation in cloud. The architecture facilitates the acquisition of satellite data, its archive and catalogue, and on demand processing for distribution to end users through a web service.
The cloud architecture is constituted of two main layers:
i) Layer 1 involves the basic satellite imagery services. It acquires satellite raw data; processes it to obtain a calibrated, geo-referenced and ortho-rectified images; and stores the processed data. It is constituted of the following components:
a. Cloud Orchestrator: manages the ingestion of satellite data in the cloud, the IaaS and the Archive and Catalogue module.
b. IaaS module: creates elastic processing chains on demand to take advantage of all resources that cloud platform provides.
c. Archive and Catalogue module: stores the images and its required metadata. It also provides standardized CSW service for layer 2, which computes and provides high added value services. It is based on the open source software GeoServer and its CSW plugin.
ii) Layer 2 offers high added value services (including post-processing and analysis of imagery data) to end users through web services. It is constituted of the Image Distribution and Visualization module (IDV): it is based on current open source geospatial stack and infrastructures (GDAL, Geoserver, GeoWebCache). It is constituted of the following components:
a. Interface with the Archive and Catalogue through a CSW service.
b. Datastore module whose function is to additionally process and optimize the images stored in the Archive and Catalogue.
c. EO server, which implements OGC Web Services
d. Tiles cache, which cache map tiles coming from a variety of sources such as OGC Web Map Service (WMS).
We present the overall design of the cloud architecture to provide EO high added value services such as emergency situations in catastrophes and natural disasters. It is implemented and tested in the free BonFIRE cloud infrastructure (http://www.bonfire-project.eu/). Making use of the BonFIRE facilities, we test massive EO data acquisition, processing, storage, distribution and service delivery using future internet technologies to demonstrate the performance of the architecture proposed.
This work is part of the Fed4FIRE European Project (number 318389) (http://www.fed4fire.eu/).

Author: 

Jonathan Becedas, Jonathan Becedas, Rubén Pérez, Gerardo González, Félix Pedrera. Manuel José Latorre

Presenter Biography: 

Dr. Jonathan Becedas is Electrical Engineer and Industrial Engineer graduated with honours. He holds a Master in Research and a PhD in Mechatronics by the Univ. of Castilla-La Mancha (UCLM), Spain. He is author of over 40 fully refereed papers in ICT and SPACE and has participated in more than 10 research projects. He held a position of Researcher and Associate Prof. in UCLM (2002-2009), and Research Associate in the University of Leicester, UK (2009-2011). He was head of the Robotics Division in Ixion Industry and Aerospace (2011-2013). He is currently R&D Manager in Elecnor Deimos where he coordinates R&D activities and projects.

Mr. Rubén Pérez Pascual is a Computer Engineer by the Univ. of Castilla La Mancha (UCLM), Spain. He is currently an R&D internship researcher in Elecnor Deimos since February 2014, where he finishes his Master Thesis about Cloud Computing, Future Internet and Federated Platforms applied to Earth Observation. He is participating in the FP7 Fed4FIRE project.

Mr. Gerardo González is Aeronautical Engineer by the Universidad Politécnica de Madrid. He is contributor of “Innovative Ideas for Micro/Nano-satellite Missions” edited by Sandau R. et al.. He holds an internship in Elecnor Deimos since February 2013, where he carried out his Master Thesis in the design of Earth Observation Satellite Missions, achieved with honours. He is working as systems engineer in the Deimos-2 project and in the R&D FP7 Fed4FIRE project.

Mr. Félix Pedrera received his MSc (2002) in Physics from the Universidad Autónoma de Madrid. He is Project Manager at Elecnor Deimos and counts more than 7 years of experience developing and coordinating Mobile, Web and research projects. He is member of the Spanish Open Knowledge Foundation Local Group, promoting environmental open data publishing at local level.

Mr. Manuel José Latorre is a Telecommunications Engineer (Electronic & Telematic specialities) by the University of Málaga (UMA). He is Senior Engineer at Elecnor Deimos (ICT Solutions division) with more than 5 years of experience as lead developer, technical director and coordinator in Augmented/Virtual Reality, Mobile, Web and research projects, such as “España Virtual” or “Arid-Lap” R&D projects.

? Top