Image information mining in time series
Extension of system functions for display analysis of satellite images
Project duration: 2007 to 2008
Project manager: Martin Pačes
The main objective of the ESA/IIM-TS project was to extend the functions of current system for satellite image information mining (KEO - Knowledge-centred Earth Observation), which is used in ESRIN, by processing and analysis of images in time series. These methods are especially useful for change detection with many practical applications, eg. assessment of growth and health of crop or identification and assessment of damage in aftermaths of natural disasters like floods, fires or extreme droughts.
Because the image information mining, especially when extended for multi-temporal analysis, deals with large amount of data, it requires high-performance computing methodologies such as algorithm parallelization together with employing computer clusters or using distributed GRID computing. The main objectives for Iguassu in the project were to find the state-of-the-art applications, methods, and algorithms for detection and analysis of meaningful features in time series of satellite images and to study different algorithm parallelization methods for KEO hardware infrastructure, GRID at the middleware level or G-POD (GRID Processing On Demand). The outputs of the project were prototypes of parallelized algorithms comparing individual images to each other, which were later used by the team of KEO developers.
Iguassu Software Systems (ISS)
The company was established in 1994 and focused on high-tech solutions including real-time software development from industrial systems to satellite control and satellite navigation. Customers include the European Space Agency and its suppliers, Eumetsat, HP, Agilent, Ingersoll-Rand and Inter-American Development Bank. ISS is currently working on:
- Parallel Data-Mining Components (prime Iguassu)
- Real-time EGNOS Performance Monitoring (prime Iguassu)
- Interference Monitoring System for GNSS Reference Stations (subcontractor of Astrium D)
- Open-standard On-line Observation Services (subcontractor of EOX)
What would you name as main benefits of the project to you and your company?
“This project was exceptional for us, because we had a chance to participate in an international consortium winning a regular ESA Invitation to tender. A significant professional benefit of this project was the opportunity to participate on development of advanced remote sensing image information mining technique. I also value the personal experience of working in the international team of the company ACS in Rome, which has more than thirty years experience of working on ESA projects.”