Published on April 3rd, 2013 | by EJC0
Featured Tool: Tomnod
1) In a nutshell
- Tool: Tomnod, meaning “Big Eye” in Mongolian, is a platform that extracts key points from satellite imagery by leveraging crowdsourcing and machine learning. By leveraging current and timely satellite imagery and fast analysis, the Tomnod system is able to find key areas of interest to help respond.
- Information: Tomnod’s platform uses geospatial databases to collect the crowd’s input and produces reliable maps from the abundance of satellite imagery.
- Data: The raw data is derived from satellite and aerial imagery where the analysis is done by the participating crowd that is recruited by Tomnod.
- Technology: Tomnod makes use of several different technologies to assess and gather the data. SatEnhance increases the resolution of satellite images; CrowdRank is an algorithm that ranks the reliability and accuracy of individuals in the crowd; through a combination of cognitive power of the human mind and the speed of computer machine learning; and the TipQ system that makes the key areas in imagery easily accessible via KMZ, email, or even a live updating Twitter account.
- Accessibility: The Tomnod system was designed from the groundup to be extremely accessible:
- The crowdportal that allows users to view imagery and contribute analysis points is accessible in any browser. Furthermore, an embeddable version is available that can be embedded in any webpage.
- The identified key locations and associated imagery are available via a number of easy to use data formats including email, Twitter, KMZ and JPG.
2) Case study
After the 2011 Christchurch earthquake, the Tomnod platform was used to crowdsource damage after the disaster to produce a map of damage in the region used to aid in the recovery effort. Similarly, a crowd of crisis mappers are identifying the flood zones and damage after Hurricane Sandy to aid in the .relief and now recovery efforts.
In July of 2012, two hikers went missing in the Peruvian Andes. Satellite imagery of the region was analysed and crowdsourced in order to assist the rescue team.
3) Media use
Using Tomnod, around 10,000 citizen scientists have contributed to the search for the tomb of Genghis Khan in Mongolia. In the Valley of the Khans Project, which was set up by National Geographic explorers, they carried out 2 million analyses to map the region where the tomb (which was lost 800 years ago) is thought to be. By tagging roads, rivers and modern or ancient structures onto the map, the citizen scientists were able to point the actual archeologists in the right direction. The mapping technology and use of digital tools allow for the scientists and their supporters to explore the region, without damaging the archeological sites.
4) What’s in it for journalists?
- The maps are geospatial, mostly making use of satellite and aerial imagery. Since these technologies generate a non-stop output, the maps are almost real-time.
- Tomnod gives a synthesis of geo-pictures, computer-generated data and human-generated data and insights.
- Thanks to the CrowdRank © technology that is part of Tomnod, its maps represent reliable information as algorithms rank the accuracy of individuals in the crowd who contribute to the analysis.
- If you want to include imagery of key areas during a live event, this is the solution.