Linda Poon is a staff writer at CityLab covering science and urban technology, including smart cities and climate change. She previously covered global health and development for NPR’s Goats and Soda blog.
Researchers are tapping into photo-sharing platforms like Instagram to analyze everything from air pollution to street design.
It’s not only your friends and family who follow your online selfies and group photos. Scientists are starting to look at them, too, though they’re more interested in what’s around you. In bulk, photos can reveal weather patterns across multiple locations, air quality of a place over time, the dynamics of a neighborhood—all sorts of information that helps researchers study cities.
At the Nanyang Technological University in Singapore, a research group is using crowdsourced photos to create a low-cost alternative to air-pollution sensors. Called AirTick, the smartphone app they’ve designed will collect photos from users and analyze how hazy the environment looks. It’ll then check each image against official air quality data, and through machine-learning the app will eventually be able to predict pollution levels based on an image alone.
AirTick creator Pan Zhengziang said in a promotional video last month that the growing concern among the public over air quality can make programs like this a success—especially in Southeast Asia, where smog has gotten so bad that governments have had to shut down schools and suspend outdoor activities. “In Singapore’s recent haze episode, around 250,000 people [have] shared their concerns via Twitter,” he said. “This has made crowdsourcing-based air quality monitoring a possibility.”
The app was tested in a 100-person study in November, according to the New Scientist, and results so far show that that it can analyze haze with 90 percent accuracy during the day. When AirTick rolls out later this year, the team hopes the app can provide the public with air pollution readings of a specific area in real time, based solely on crowdsourced photos. If the levels indicate that the air quality is hazardous, the app will tell them the nearest place to get medical attention.
It can also be helpful for urban planners looking to improve clean energy by installing solar panels. In the video, Pan points to the app’s “hidden gem”—its ability to track the direction, duration, and intensity of sunlight from the photos. “Based on this insight, I am developing optimization techniques to help urban planners determine not only the location but also the orientations of solar panels for deployment,” he said.
Though it may sound like an overly simplistic solution to one of the world’s most complex problems, the AirTick app addresses a larger issue with data collection. As CityLab previously reported, even the wealthiest cities have little understanding of air pollution patterns because traditional sensors don’t track daily exposure. Instead, many depend on satellite data. That’s starting to change as researchers like Pan look to personal or portable air sensors to crowdsource data for environmental studies.
The many practical uses of shared photos
Pan is among a small handful of scientists to use something so readily available: photos. New Scientist reports that in 2015 alone, between 2 and 3 trillion photos were uploaded to the Internet through social media and messaging apps by roughly 2 billion smartphone users. In Singapore, with a population of 5.4 million, about nine in 10 people have access to a smartphone, according to a survey by Deloitte.
In 2012, a professor at City University of New York made headlines when he analyzed the “selfie style” of five cities using hundreds of photos taken from public Instagram accounts. (He added a sixth city, London, in 2015.) The cheeky study used data to determine how happy people were in each city, but the real objective, as CityLab reported, was to illustrate how new digital media can be analyzed and turn into a research tools for scientists.
Indeed, photo-sharing platforms offer researchers a seemingly endless database of visuals. A quick search of the term “Instagram” on Google Scholar pulls up dozens of studies on everything from health to feminism to museum experience. In one study from 2013, researchers used Instagram and and Foursquare to look at city dynamics and social behavior in the urban environment. In another study, researchers mined more than 2.3 million Instagram photos from 13 cities to track cultural changes around the world.
Further south, Peru and the U.S. Agency for International Development are encouraging Peruvians to use their smartphone cameras to track the country’s alarming waste problem. The ongoing campaign, which enlists the help of vultures with cameras and GPS strapped on to them, uses geographic coordinates from the photos to plot images onto a live map showing just how widespread the problem is.
Among the most widely used photos so far are the ones generated by Google Street View. They’ve been used to study the effect of neighborhoods on children’s health, street designs that are most dangerous for pedestrians, and sidewalk accessibility. As the public’s digital presence grows, in the form of photos, tweets, and “check-ins,” this information is becoming more useful and accessible for researchers around the world.