Crowdsourced Satellite Image Analysis

There are many circumstances where the amount of data vastly exceeds the ability to process and analyze it—and computers can only do so much. Enter crowdsourcing. Steve Coast points to Digital Globe’s Tomnod project, which basically crowdsources satellite image analysis. In the case of the current project to  map the presence of Weddell seals on the Antarctic Peninsula and the ice floes of the Weddell Sea, users are given an image tile and asked to indicate whether there are seals in the image. It’s harder than it looks, but it’s the kind of routine task that most people can do—many hands, light work and all that—and it helps researchers focus their attention where it needs focusing. (A similar campaign for the Ross Sea took place in 2016.)

Another ongoing campaign asks users to identify flooded and damaged infrastructure and trash heaps in post-Hurricane Maria Puerto Rico.

Volunteers Mapping Post-Hurricane Puerto Rico

When disaster strikes, crowdmapping kicks into high gear. Last Friday, six universities hosted mapathons where volunteers, using satellite imagery, contributed to the map of Puerto Rico and other hurricane-damaged areas on OpenStreetMap. More from one of the universities involved. Here’s the relevant project page on the OSM Wiki.

Esri Makes Satellite Imagery Available to OpenStreetMap Editors

Esri is making its satellite imagery collection available to OpenStreetMap editors.

Today Esri is proud to announce that we are making our own global collection of satellite imagery available to the OSM community directly through our existing World Imagery Service. This regularly updated resource provides one meter or better satellite and aerial photography in many parts of the world, 15m TerraColor imagery at small and mid-scales (~1:591M down to ~1:72k), 2.5m SPOT Imagery (~1:288k to ~1:72k), 1 meter or better NAIP in the US and many other curated sources, so we know it will make a welcome addition to OSM’s growing catalog of reference layers.

OSM editors have been able to trace maps from satellite imagery for years; other sources of such imagery have included Bing and Yahoo (back when Yahoo Maps was a thing). Different sources have different strengths, so this can only help the project. (Esri’s imagery makes no difference where I am, but that’s not a surprise.)

Google Map Maker Is Now Closed

Google Map Maker, Google’s tool to allow users to edit its maps, has been shut down, Ars Technica reports. “A support page went up over the weekend declaring that Map Maker is closed but that ‘many of its features are being integrated into Google Maps.’” You may recall that Map Maker was temporarily suspended in 2015 after a series of embarrassing edits came to light; its editing tools have been increasingly limited to a smaller circle of editors.

Mapping Safe Washrooms

In response to measures like North Carolina’s House Bill 2, which restricts access to public washrooms by transgender people, crowdsourced online maps of safe washrooms—places with unisex or gender-neutral washrooms, or that let transgender people use the washroom that matches their gender identity—have been created: Refuge Restrooms has both a list and a map view; Safe Bathrooms uses Google My Maps. These maps seem like the modern-day equivalent of The Negro Motorist Green Book for trans people. [WMS]

Missing Maps

Quartz takes a look at the Missing Maps project, which I suppose can best be described as a way to jumpstart mapping the unmapped developing regions of the world with OpenStreetMap. What’s interesting about Missing Maps is how it systematically deals out tasks to people best able to do them: remote volunteers trace imagery, community volunteers do the tagging and labelling. There’s even an app, MapSwipe, that gives its users “the ability to swipe through satellite images and indicate if they contain features like houses, roads or paths. These are then forwarded onto Missing Maps for precise marking of these features.” [WMS]

Putting Slums on the Map

kibera

In Smithsonian, Erin Blakemore explores the on-the-ground, amateur efforts to get disadvantaged communities—slums, shanty towns, whatever they may be called—on the map, like the Map Kibera and Mapillary projects, and the implications of such projects.

Sterling Quinn, who is earning his Ph.D. in geography at Penn State, notes that there are downsides to user-generated maps. Just because an underserved community makes its way onto the map doesn’t mean it becomes less vulnerable, says Sterling. “Putting yourself on the map may make you more vulnerable to people who want to exploit the area,” he tells Smithsonian.com.

[Dave Smith]

Previously: The Geospatial Revolution Project, Episode FourCrowdsourcing Street Photos of Dar es Salaam.

Crowdsourcing Street Photos of Dar es Salaam

Point Google Maps or OpenStreetMap at a city like Dar es Salaam, Tanzania and you’ll get a reasonably good map. What you won’t get is Street View or street-level imagery—or, necessarily, the data that comes from a street-level understanding of the territory. NPR’s Nadia Whitehead looks at a joint project of the World Bank and Mapillary, a company that crowdsources street-level photos, to produce those images. “Volunteers are mounting camera rigs to their tuk tuks—three-wheeled motor-powered vehicles—to snap pictures as they cruise Dar es Salaam’s dirt roads. Others download the Mapillary app on their smartphones and capture images as they walk or hitch rides on motorbikes. In all, more than 260 people have volunteered.” [via]

Mapping Swiss German Dialects

swiss-german-app

Researchers are mapping the shift in Swiss German dialect usage via an iOS app. The app asks users to take a 16-question survey based on maps from a language atlas that mapped Swiss German usage circa 1950. The app predicts the user’s actual home dialect location based on those maps; differences between that prediction and the user’s actual home dialect location reveal how Swiss German has changed over time. They ended up getting responses from 60,000 speakers. PLOS ONE article. [via]

And in Google Maps News …

Google’s Map Maker is in the process of reopening, with six countries reopening on August 10 and another 45 countries last Monday. Map Maker, Google’s tool allowing users to make changes to Google Maps, was suspended last May after some embarrassing edits came to light. Regional leads are now in place to review user edits before they go live on the map.

If mapcodes and other geographical shortcodes aren’t Googly enough for you, take a look at Open Location Codes, a Google-developed, open-sourced project. Generated algorithmically rather than with data tables. Announced for developers last April, they can now be used in Google Maps searches.

Google Maps Edits Cause Embarrassment

Some embarrassment for Google Maps last week, as they were forced to apologize for an image of the Android mascot peeing on an Apple logo that turned up on the map near Rawalpindi in Pakistan. To say nothing of the phrase “Google review policy is crap” etched into nearby Takht Pari Forest. Both have since been removed. Boing Boing, the Guardian, The Verge.

To be fair to Google, crowdsourcing map data does have its pitfalls: OpenStreetMap has to deal with this sort of thing all the time. You have to have something in place to deal with bad-faith edits. None of the edits I’ve made to Google Maps went through without someone reviewing them, so I’m surprised that this could happen. That said, when you need your map updated fast (such as during natural disasters like yesterday’s earthquake in Nepal), it’s hard to beat crowdsourcing.

As always, it’s important to keep in mind that all online maps have their shortcomings.

OpenStreetMap’s New Map Editor

OpenStreetMap has launched a new map editing interface that runs, for the first time, in HTML5. (Potlatch, the previous web-based map editor, uses Flash, and JOSM runs in Java, which I always thought was ironic for an open project.) The editor, called iD, is live now, and is designed to make editing the map more accessible to beginning mapmakers. I’ve given it a quick try this morning. My summary judgment is that if you have any experience using another editor, you should stick with it. iD is far slower than Potlatch at the moment, and does things sufficiently differently that you might have a hard time finding things. I made a mess trying to edit the existing map. But will it lower the barrier to making new contributions, particularly for casual or non-technical contributors? I hope so.