Project Home
Digital Puglia (SARA)
InSAR
Interferometric
Synthetic Aperture
Radar
GIOD
Globally Interconnected Object Databases
LIGO
Laser Interferometer Gravitational-Wave Observatory)
XSIL
Extensible Scientific Interchange Language
|
Digital Puglia
SARA (Synthetic Aperture Radar Atlas) is a web-based retrieval mechanism that has been online for over a year at the University of Lecce in Italy, at Caltech in California, and at SDSC in San Diego, California. The opening page presents a user with a picture of the Earth, and clicking on it can zoom in or out, depending on how the rest of the form is set. More distant views show coasts, countries, and rivers; closer views also show roads, railways, and city names. Also shown on the map are rectangular 'tracks' indicating that a SAR image is associated with that region. The user may select a particular track from the map by clicking on it; if multiple tracks contain the chosen point then the user is asked to select from a list of those tracks.
When a track has been selected, a Java applet appears, with a thumbnail image of the track area (perhaps 300 pixels wide by 1000 high), together with other controls that allow selection of a subset of the full track, an output format, and the false-coloring information. Since there are several SARA data servers around the world, the user can also select which of these should be used to create and retrieve the data. When the data-order is complete, the user can download the resulting image.
Digital Puglia is a new project, emphasizing not only retrieval of data to the client's workstation, but also customized processing of the data. It may be that the data is too large for downloading to be practical, or the client may not have the relevant processing software. The processing at the server may be as simple as a change of format, or it may be that a user does compute-intensive image processing such as principal component analysis, supervised classification, or pattern matching. By the term "procession on demand", we mean a data archive connected to a powerful compute server at high bandwidth, controlled by a client who may be connected at low bandwidth.
Processing on demand may also be necessary to get timely access to recent data; relevant to a hazard situation such as fire or earthquake, or to monitor a potential hazard such as an oil slick. In the case of on-demand SAR processing, the user interface should allow control of doppler parameters, chirp generation methods, and other parameters that control the processing software.
Another use for the active library is that there may be potential knowledge to be gained through combining and fusing data objects. In the Digital Puglia project, we will concentrate on multitemporal imaging, where the color channels of an image represent the surface of the Earth at different times, rather than different wavelengths or polarizations. In general, the collection of (monochrome) images from different times will not be geo-referenced: thus this is the major computational task. Once we have an infrastructure for creating multitemporal images, it is a short step to creating interferometric images, showing ground motion at the centimeter level.
A Presentation by Roy Williams on Digital Puglia
Visualization and Processing of Multichannel Images
The Interferometric SAR project at CACR
More information about the Digital Puglia project and its antecedent, SARA, may be found in the following papers. Most papers are in PDF Format. Download the PDF Reader here.
The Digital Puglia Project: An Active Digital LIbrary of Remote Sensing Data
Giovanni Aloisio, Massimo Cafaro, Roy Williams
Proceedings of the 7th International Conference on High Performance Computing and
Networking Europe. April 12 - 14, 1999. Amsterdam, The Netherlands, (Springer Lecture
Notes in Computer Science vol 1593, page 563).
PDF
An XML Architecture for High-Performance Web-Based Analysis of Remote-Sensing
Archives
Giovanni Aloisio, Giovanni Milillo, and Roy Williams
September 1998
Submitted to Future Generation Computing Systems
pdf.
Interfaces to Scientific Data Archives
Report of the NSF Workshop
Roy Williams, Julian Bunn, Reagan Moore and James Pool,
May 1998.
Choose pdf or HTML.
A High-Performance Active Digital Library
Roy Williams and Bruce Sears
Proceedings of HPCN98, Amsterdam, April 1998, eds. L. O. Herzberger and P. M. A. Sloot,
Lect. Notes Comp. Sci. (Springer)
pdf
A Distributed Web-based Metacomputing Environment
G. Aloisio, M. Cafaro, P. Messina, R.D. Williams,
Proceedings of HPCN97, Vienna, April 1997, eds. L.O. Herzberger and P.M.A. Sloot, Lect.
Notes Comp. Sci. (Springer)
pdf