The Open Aerial Map architecture is based on distributed services and user contributed content. OAM will implement international geospatial and W3C standards in order to create a scalable, interoperable and open source system.
Only free data will be available and publishable through OAM. The map will consist of user contributed imagery collected with traditional remote sensing systems (satellites and aircrafts) and newly developed low cost aerial imaging systems (small sensor mounted on UAVs, remote controlled planes, blimps, kites, etc).
In line with the concept of OpenStreetMap, OAM aims to create, display, and distribute free geographic data that will be available in geospatial formats, as static images, and through an API.
Manual Image Registering
The first OAM upload interface will consist of an online image rectifying application, allowing contributors to upload single images and reference them to existing geospatial layers. This will be very similar to geotagging photos in Flickr or Google Earth, by manually placing each image in the correct position on the Earth surface.
The Map Rectifier from Metacarta Labs is a good example of how the single image uploader would work. Each georeferenced image will merge into the global OAM mosaic and its metadata entry created in a daily generated GeoRSS/Feature catalog feed.
Geo-Imagery and Services
Any dataset already containing geo-referencing information in standard GIS formats (e.g. geotiff, world file, img) will be integrated automatically. Imagery uploaded to OAM will be integrated into the global mosaic using a catalog-like system. If reliability or bandwidth consistency becomes a problem, OAM will be hosted on dedicated servers.
No real time processing will be included as part of OAM functionalities at this stage. Manual distortion correction, contrast and color enhancement, croping, etc will need to be done before uploading. Data will be mosaicked in a catalog fashion to update the world map on a regular time interval.
Upon upload a number of image characteristics will be defined by the user, including at minimum: Acquisition time and date, sensor, colors, and notable anomalies. This will allow the sorting of imagery by these parameters. More extensive meta data tags will also be permitted.
Automatic Image Georegistration
Image files with single frame center/corner control points will be uploaded to OAM and automatically merged into the global mosaic using a pattern recognition approached similar to that used by current automatic panorama stitching like autostitch. This approach will eliminate the need for a manual referencing interface, but is not intended to phase out the upload of referenced imagery.
Each image added to the world map will be retained indefinitely, to permit the storage of a time series of data.