|
Titel |
The iMars WebGIS – Spatio-Temporal Data Queries and Single Image Map Web Services |
VerfasserIn |
Sebastian Walter, Ralf Steikert, Bjoern Schreiner, Jan-Peter Muller, Stephan van Gasselt, Panagiotis Sidiropoulos, Julia Lanz-Kroechert |
Konferenz |
EGU General Assembly 2017
|
Medientyp |
Artikel
|
Sprache |
en
|
Digitales Dokument |
PDF |
Erschienen |
In: GRA - Volume 19 (2017) |
Datensatznummer |
250154115
|
Publikation (Nr.) |
EGU/EGU2017-19171.pdf |
|
|
|
Zusammenfassung |
Introduction: Web-based planetary image dissemination platforms usually show
outline coverages of the data and offer querying for metadata as well as preview
and download, e.g. the HRSC Mapserver (Walter & van Gasselt, 2014). Here we
introduce a new approach for a system dedicated to change detection by simultanous
visualisation of single-image time series in a multi-temporal context. While the usual form
of presenting multi-orbit datasets is the merge of the data into a larger mosaic,
we want to stay with the single image as an important snapshot of the planetary
surface at a specific time. In the context of the EU FP-7 iMars project we process and
ingest vast amounts of automatically co-registered (ACRO) images. The base of the
co-registration are the high precision HRSC multi-orbit quadrangle image mosaics,
which are based on bundle-block-adjusted multi-orbit HRSC DTMs. Additionally
we make use of the existing bundle-adjusted HRSC single images available at the
PDS archives. A prototype demonstrating the presented features is available at
http://imars.planet.fu-berlin.de.
Multi-temporal database: In order to locate multiple coverage of images and select
images based on spatio-temporal queries, we converge available coverage catalogs for various
NASA imaging missions into a relational database management system with geometry
support. We harvest available metadata entries during our processing pipeline using the
Integrated Software for Imagers and Spectrometers (ISIS) software. Currently, this database
contains image outlines from the MGS/MOC, MRO/CTX and the MO/THEMIS instruments
with imaging dates ranging from 1996 to the present. For the MEx/HRSC data, we already
maintain a database which we automatically update with custom software based on the
VICAR environment.
Web Map Service with time support: The MapServer software is connected to the
database and provides Web Map Services (WMS) with time support based on the
START_TIME image attribute. It allows temporal WMS GetMap requests by setting
additional TIME parameter values in the request. The values for the parameter represent an
interval defined by its lower and upper bounds. As the WMS time standard only supports one
time variable, only the start times of the images are considered. If no time values
are submitted with the request, the full time range of all images is assumed as the
default.
Dynamic single image WMS: To compare images from different acquisition times at
sites of multiple coverage, we have to load every image as a single WMS layer. Due to the
vast amount of single images we need a way to set up the layers in a dynamic way
– the map server does not know the images to be served beforehand. We use the
MapScript interface to dynamically access MapServer’s objects and configure the file
name and path of the requested image in the map configuration. The layers are
created on-the-fly each representing only one single image. On the frontend side, the
vendor-specific WMS request parameter (PRODUCTID) has to be appended to the
regular set of WMS parameters. The request is then passed on to the MapScript
instance.
Web Map Tile Cache: In order to speed up access of the WMS requests, a MapCache
instance has been integrated in the pipeline. As it is not aware of the available PDS product
IDs which will be queried, the PRODUCTID parameter is configured as an additional
dimension of the cache. The WMS request is received by the Apache webserver configured
with the MapCache module. If the tile is available in the tile cache, it is immediately
commited to the client. If not available, the tile request is forwarded to Apache and the
MapScript module. The Python script intercepts the WMS request and extracts the product ID
from the parameter chain. It loads the layer object from the map file and appends
the file name and path of the inquired image. After some possible further image
processing inside the script (stretching, color matching), the request is submitted to the
MapServer backend which in turn delivers the response back to the MapCache
instance.
Web frontend: We have implemented a web-GIS frontend based on various
OpenLayers components. The basemap is a global color-hillshaded HRSC bundle-adjusted
DTM mosaic with a resolution of 50 m per pixel. The new bundle-block-adjusted
qudrangle mosaics of the MC-11 quadrangle, both image and DTM, are included
with opacity slider options. The layer user interface has been adapted on the base
of the ol3-layerswitcher and extended by foldable and switchable groups, layer
sorting (by resolution, by time and alphabeticallly) and reordering (drag-and-drop).
A collapsible time panel accomodates a time slider interface where the user can
filter the visible data by a range of Mars or Earth dates and/or by solar longitudes.
The visualisation of time-series of single images is controlled by a specific toolbar
enabling the workflow of image selection (by point or bounding box), dynamic image
loading and playback of single images in a video player-like environment. During a
stress-test campaign we could demonstrate that the system is capable of serving up
to 10 simultaneous users on its current lightweight development hardware. It is
planned to relocate the software to more powerful hardware by the time of this
conference.
Conclusions/Outlook: The iMars webGIS is an expert tool for the detection and
visualization of surface changes. We demonstrate a technique to dynamically retrieve and
display single images based on the time-series structure of the data. Together with
the multi-temporal database and its MapServer/MapCache backend it provides a
stable and high performance environment for the dissemination of the various iMars
products.
Acknowledgements: This research has received funding from the EU’s FP7 Programme
under iMars 607379 and by the German Space Agency (DLR Bonn), grant 50 QM 1301
(HRSC on Mars Express). |
|
|
|
|
|