|Author:||Missing Maps Project|
|License:||Data (CC BY 3.0) and App (BSD) (free of charge)|
|Platforms:||Android and iOS|
Super-simple contribution to humanitarian mapping (indirectly) by helping find features in imagery
MapSwipe mapswipe.org is an app developed by a team in collaboration with the Missing Maps Project as an easy way for people to contribute to humanitarian mapping efforts. The contribution is indirect. A form of "second level crowdsourcing".
Users swipe through satellite images of a region, tapping the screen when they see features they’re looking for including settlements, roads and rivers. This information is fed back to mappers who need this information to build detailed and useful maps. At present, they have to spend days scrolling through thousands of images of uninhabited forest or scrubland looking for communities that need mapping. Now, members of the public can directly contribute to MSF’s medical activities by locating people in need more quickly so mappers, and ultimately medical professionals on the ground, can get straight to work.
- 1 Using MapSwipe
- 2 Requesting a MapSwipe mission
- 3 Development
- 4 Blogs and News
Download the (free) app
- The app store: https://itunes.apple.com/gb/app/mapswipe/id1133855392?mt=8
- The Google play store: https://play.google.com/store/apps/details?id=org.missingmaps.mapswipe
- The newest release from Github: https://github.com/mapswipe/mapswipe/releases
How to use MapSwipe
MapSwipe is very intuitive. Follow next steps:
- Choose your 'mission'. Each mission will indicate what you are looking for (for example, buildings and roads).
- Swipe left to see the next tile.
- Tap once on a square to identify a feature. The square will turn green.
- Tap twice if you are not sure, but you think there is a feature. The square will turn orange.
- Tap three times to identify bad / no imagery or cloud cover. The square will turn red.
Need some help?
Check out next resources:
- MapSwipe tutorial
- MapSwipe guide for more help on feature identification.
- See live edits - displays live Mapswipe contributions inspired by https://wiki.openstreetmap.org/wiki/Show_Me_The_Way.
We welcome feedback. Please leave your questions, comments, or concerns on this form. MapSwipe is an open source project created and maintained by volunteers like you, so the best way to see new features implemented is to get involved and make it happen. More technical feedback can be added to https://github.com/mapswipe
Requesting a MapSwipe mission
Create your project
MapSwipe is an app developed within the Missing Maps Project, but any organisation or local OSM community can request a MapSwipe mission.
The different project types you can request are:
- Build area : A 6 squares layout is used for this project type. By tapping MapSwipe users can classify a tile of satellite imagery as yes, maybe or bad_imagery. Project managers can define which objects to look for, e.g. "buildings". Furthermore, they can specify the tile server of the background satellite imagery, e.g. "bing" or a custom tileserver. By scanning imagery on buildings, the result data set will give an overview of the population distribution.
- Footprint (upcoming)
- Change detection (upcoming)
Find all information to request a MapSwipe mission in this document.
Follow progress and engage
You can follow the progress of your task in the app itself, or on MapSwipe analytics created by the Heidelberg University team.
Download and work with the data
Once your task is at 100%, the data is available to download on MapSwipe analytics, and is being used either directly or to guide further mapping priorities. This for example feeding into the design of OSM Tasking Manager projects, in an manual observational way, or via automated project designing mechanisms which are still somewhat under development.
The GIScience Resarch Groups at Heidelberg University provides python scripts to process the mapswipe data: https://mapswipe.heigit.org
You can join the MapSwipe conversations on Slack. Please do sign up here: https://mapswipe.slack.com/
Open Source code on github
The app is hosted on https://firebase.google.com and it is managed and developed by the MapSwipe volunteer working group that meets every two weeks. If you would like to be part of these discussions.
- App (react-native): https://github.com/mapswipe/mapswipe
- Back End (python): https://github.com/mapswipe/python-mapswipe-workers
Provides statistics, visualisation and data for MapSwipe:
- Analytics: Show all MapSwipe projects on a map. https://github.com/GIScience/mapswipe-analytics-backend, https://github.com/GIScience/mapswipe-analytics-client
- Live: Show the latest MapSwipe results. https://github.com/GIScience/mapswipe-live
- Tutorial: Provide some simple examples. https://github.com/GIScience/mapswipe-tutorial
- Processing: Get MapSwipe data ready to use in a GIS program. https://github.com/GIScience/mapswipe-processing
MapSwipe and Machine Learning
A suite of tools for doing machine learning with the data from Mapswipe: https://github.com/philiptromans/mapswipe-ml-dataset-generator#
Blogs and News
- Saving lives with social media: The power of the crowd in emergencies - https://www.dw.com/en/saving-lives-with-social-media-the-power-of-the-crowd-in-emergencies/a-48305661
- MapSwipe - the story continues http://www.missingmaps.org/blog/2018/06/25/mapswipe-story/
- A Mapping Interface That Saves Lives - https://www.fastcompany.com/90138077/a-mapping-interface-that-saves-lives
- MSF Scientific day presentation by Herfort, B., Reinmuth
- MapSwipe Dashboard - https://disastermappers.wordpress.com/2017/03/28/10-million-contributions-its-time-for-mapswipe-analytics/
- MapSwipe - https://helenaroth.com/mapswipe/
- MapSwipe: Create Crisis Maps While Waiting for the Bus - http://sm4good.com/2016/12/02/mapswipe-create-crisis-maps-waiting-bus/
- To Map Crisis Zones, Humanitarian Groups Turn to the Crowd - https://www.citylab.com/life/2016/10/mapswipe-map-crisis-zones-conflict-regions-red-cross/505376/
- App that helps aid workers: 'I watched TV while mapping 100km of Nigeria' - https://www.theguardian.com/global-development/2016/aug/05/app-that-helps-aid-workers-i-watched-tv-while-mapping-100km-of-nigeria