Professor Uwe Sörgel
Institute for Photogrammetry and Geoinformatics, University of Stuttgart
Presentation: "Paradigm Changes and Trends in Remote Sensing"
We are currently experiencing very dynamic developments in the field of remote sensing, which present us with new challenges but also open up opportunities. In the past, satellite remote sensing was operated by governments or supranational organizations, there were only a few operational satellites at a time, which were very expensive.
Today, more and more private companies are coming into play, launching entire constellations of smaller and cheaper units into orbit, dramatically improving data availability and timeliness. This also means that their analysis can no longer be carried out by experts alone. Rather, more and more lay people are involved, for example in the form of paid crowdsourcing for the annotation of remote sensing data, for instance, used as training data for land cover classification. Finally, remote sensing-based services are also offered to the public, such as the European Ground Motion Service.
Professor Charles Toth
Dept. of Civil, Environmental and Geodetic Engineering, The Ohio State University
Presentation: "Mobile Mapping for Geoinformation"
Mobile mapping is a technique used to gather geographical information, such as the location of natural landmarks and man-made objects, from a moving vehicle. The technology has been around for decades, but recent advances in sensor technologies and computers have made mobile mapping easier, affordable, and more accurate. With the introduction of smart devices that can collect huge volume of location data and increasingly visual information, crowdsourcing is emerging as a new source of geospatial information. More recently, assisted, and autonomous vehicle technologies are becoming an excellent source to acquire visual data in urban environment, offering an inexpensive alternative to traditional mobile mapping. Currently, most of the geoinformation acquired by autonomous vehicles are discarded, but as communication technologies continue to improve and data analytics technologies supported by cloud processing are becoming widely available, these platforms are expected to emerge as the primary source for geoinformation for road mapping, generating HD maps, traffic monitoring, etc.
This presentation first looks into the sensing aspects of mobile mapping technologies. Then, the key characteristics of crowdsourced data are reviewed, including the processing aspects, such as the deep learning components. Finally, the potential of crowdsensing to acquire geospatial data along transportation corridors and cities are discussed.