• Subscribe

Eye in the sky

Nearly 40 years ago, some scientists took a picture. This wasn’t any old snapshot — this was an image of the entire planet.

The Blue Marble photograph, snapped by the crew of the Apollo 17 spacecraft, went on to become one of the most reproduced images ever taken. It’s also among the most humbling pictures the average person will ever see.

<strong>This image of the Earth</strong> is one of the most reproduced images ever taken, and for good reason. How could you look at this image and NOT think deeply about the universe.

Everyone you’ve ever known, everyone they’ve ever known, and every single living thing we know to exist is contained in that little marble.  

Viewing the Earth from a distance tends to teach people a lot about our role in the universe. While they may inspire feelings of awe and connection to the universe, images like the Blue Marble don’t do much to teach us about life on our planet.

That said, scientists like Dr. Xiaoxiang Zhu (professor at TU Munich & German Aerospace Center and presenter at the ISC 2021 conference) are attempting to use various technologies to capture aspects of our planet’s operations.

Called Earth observation, this field uses a variety of techniques to study our world. What’s more, by utilizing all of these different technologies — from machine learning to natural language processing (NLP)  Dr. Zhu is able to learn a lot about the place we call home.

Keeping a watchful eye

As Dr. Zhu points out, Earth observation is what it sounds like — it’s the scientific field of observing the Earth’s surface. She relies on satellites, unmanned aerial vehicles, and airplanes to conduct her work. Each of these observational machines is equipped with sensors to take various measurements of the planet. 

Specifically, Dr. Zhu works with the data itself. She’s thinking of ways to more effectively use the data that is collected.

“I'm working on the informational retrieval part,” says Dr. Zhu. “This means how to mine geo-information from earth observation data to support downstream applications, such as environmental science, earth system science, climate change, and UN’s Sustainable Development Goals.”

<strong>Dr. Xiaoxiang Zhu</strong> is professor at TU Munich & German Aerospace Center and presenter at the ISC 2021 conference. By using tools like machine learning, Dr. Zhu is able to learn some interesting details about our world.

Dr. Zhu continues: “We are now living in the golden era of earth observation, being offered with tens of petabytes free and open Earth observation data, it’s no longer sufficient to use these classic approaches to evaluate them. Instead, we need artificial intelligence and machine learning more and more to help with data-driven scientific discovery.”

Of course, so much work would necessitate some kind of payoff. There are a lot of applications for earth observation, but one that Dr. Zhu mentions is with the identification of extremely poor communities. Often referred to as slums, there are a variety of ways to collect information about them.

To begin, Dr. Zhu can use a satellite to get the 3D shapes of buildings. Then, satellite hyperspectral sensors give her the information needed to estimate the roofing material. In certain cases, social media analysis can help identify a building’s function.

Just being able to locate these poor communities at all is certainly an achievement, but Dr. Zhu and her colleagues are thinking beyond that. By combining the 3D information generated by the satellite, Dr. Zhu is also able to get a transparent calculation of the population density capacity of a given area.

In these communities, population density is often underestimated. Having real data to work with can help officials better understand the true size of the poverty problem they’re facing. 

Ethics of observation

Of course, like any dedicated scientist, Dr. Zhu must constantly think about the ethics of her profession.

“We intend to close the geoinformation gap in developing areas,” says Dr. Zhu. “How can we make our models trained on data rich areas, such as Europe, transferrable to the global south?” 

Her research into poorer communities touches on these ethics. By analyzing tens of petabytes of earth observation data, Dr. Zhu’s team has put great effort into using AI and data science to derive global and quality-controlled geo-parameters of the built environment, including building models, population densities, as well as the functions of buildings. While talking about making global slum maps openly accessible, she has her concerns. 

“This (making slum maps publicly available) is, of course, research work done with very good will — we hope this geoinformation will be used for the better planning of these informal settlements,” says Dr. Zhu. “But how could you avoid fairness issues once the data is published? People who are living in these slums should not get any social disadvantage. So, this is an open question — the whole community needs joint forces to work on solutions.”

While certainly inspiring, this humanistic ethical backbone isn’t out of character for Dr. Zhu. In fact, it seems to be at the core of her scientific work.

She mentions that her love of science grew when she first saw the Blue Marble image. Like billions of other people, it seems that this picture had a lasting impact on her that has helped drive her love of science into a socially conscious direction.

“I like my research field. (Earth observation) is more or less my hobby,” says Dr. Zhu. “In particular, now I am making the information and knowledge we derived from Earth observation have social impacts.”

Join the conversation

Do you have story ideas or something to contribute? Let us know!

Copyright © 2021 Science Node ™  |  Privacy Notice  |  Sitemap

Disclaimer: While Science Node ™ does its best to provide complete and up-to-date information, it does not warrant that the information is error-free and disclaims all liability with respect to results from the use of the information.

Republish

We encourage you to republish this article online and in print, it’s free under our creative commons attribution license, but please follow some simple guidelines:
  1. You have to credit our authors.
  2. You have to credit ScienceNode.org — where possible include our logo with a link back to the original article.
  3. You can simply run the first few lines of the article and then add: “Read the full article on ScienceNode.org” containing a link back to the original article.
  4. The easiest way to get the article on your site is to embed the code below.