BRIT Aerial Surveys – New Projects Up in the Air

March 05, 2014

When we look at something from a different perspective, we usually see something new—something we might have not seen before or something familiar seen in a new context. Discovery of the new is precisely why we’ve recently started to explore ways to look at landscapes and ecosystems from a new perspective—namely, from the air. With very simple and affordable components such as helium balloons, kites, and point-and-shoot cameras, we’ve been able to capture great aerial images of the BRIT campus. And with some rather sophisticated software (details below) we were able to analyze the images to create a number of products that are valuable to our ongoing understanding of landscapes and ecosystems.

Hasn’t this already been done?

When we first posted our aerial survey imagery online through Facebook, some people wondered why we would bother doing this when comprehensive aerial imagery is available from Google and other sources. This is a great question, but we have some really good reasons for conducting our own aerial surveys.

The first and most important reason is that we can perform surveys at precise times that align with our research goals. For example, we were able to survey our prairie the day before the horses arrived and soon after they departed. This allowed us to determine the impact that the horses had on our prairie. We plan to continue periodic surveys of our prairie and other locations to be able to monitor changes in plant growth and composition over time.

Another reason we conduct our own surveys is image resolution. The satellite imagery that is typically provided by Google in urban areas is at a resolution of 2.5 meters, meaning that an object measuring 2.5 meters on the ground would measure one pixel in the image. With our own imagery, we had a resolution of 0.02 meters or 2 centimeters per pixel! In more concrete terms, that means that a U.S. quarter would be visible on the ground and would measure one pixel across in the images we captured. This kind of resolution allows us to identify plant species in many cases.

Finally, our method of capturing the landscape with multiple images from multiple locations (as the camera moves above the ground) allows us to perform some really interesting and powerful image analysis that would not be possible with Google or other imagery that is only a single, two-dimensional image. I’ll explain this briefly in the products section below and in greater detail in subsequent posts.

We started this project with a great set of tools and information from Public Lab (www.publiclab.org), an organization that has developed a number of open-source tools for environmental exploration including kits for balloon and kite mapping. Kites have a long history as systems for aerial surveys, and our initial plan was to conduct our surveys using the kites we bought through Public Lab. The problem was that the weather just didn’t always cooperate. On days when the wind isn’t blowing much, helium balloons still allow us aerial access so our first surveys were conducted with a helium balloon. However, at the end of February, we were lucky enough to have some steady winds that allowed us to perform our first kite survey with help from the Green Revolution students (details coming in a future blog post).

The Survey Process

Using the Public Lab balloon kit, some helium, and a custom-built camera rig with a basic Canon point-and-shoot camera, we performed three aerial surveys in October and November of 2013. Below are some pictures from these surveys that demonstrate the basic process.

homemade camera rig
This custom-built camera rig holds the Canon camera pointing down with a vane to keep it from spinning in the wind.

 

inflating the balloon
Inflating the balloon with helium.

 

attaching the camera
Fastening the camera to the camera rig and attach the rig to the balloon.

 

launching the balloon and camera
Launching the balloon. A string tethers the survey platform which can be raised, lowered, and moved across the landscape by the surveyor. Note that this camera rig is different from the one pictured above. This rig is more appropriate for kites but spins too much on a balloon so we made a new rig just for our balloon surveys.

 

spooling out the balloon to the survey height
Raising the balloon to the survey altitude.

 

looking down at the launch site from the balloon
One of the first aerial photos of the launch site as the balloon is raised.

 

BRIT building from the air
The camera is set to take a picture every five seconds so we were lucky to get this perfectly framed image of the BRIT building. You can see the tether on the lower right side of the image. You can also see a nice reflection of the clouds in our solar panels!

 

view of BRIT braided path
We surveyed not just the back prairie, but the entire grounds, including BRIT’s braided path gardens and the bioswales in the parking lot.

 

The Products

On our first balloon survey, we collected 378 images of the BRIT prairie. Each of these provides a high-resolution image of the prairie from a perspective we’ve never seen before. But great images aren’t all we can get from these surveys. With a variety of powerful software and analysis techniques, we’re able to generate a number of different products from these images. Below is a small sample of such products with brief descriptions, but we’ll explore these in greater detail in a future post.

 

point cloud
We use software called VisualSFM to analyze all of the images to identify features in the landscape that appear in multiple images. Each of these features is mapped as a single point, and when all the images and points are combined, the software generates a three-dimensional “point cloud” that represents the location of not only each feature, but the location of the camera when each image was taken. This image shows the point cloud of the BRIT campus and the many locations of the camera during the 100 meter and 200 meter surveys.

 

high resolution
With all of the common features discovered within all the pictures, the software is able to “stitch” all of the images together into one high-resolution image. This is the image for which one could “zoom” all the way in to spot a single coin on the ground!

 

BRIT Digital  Elevation Model
Since the image analysis also produced 3D data of the landscape, we are able to generate a digital elevation model (DEM) which represents the elevation of the landscape, trees, and structures in different shades of gray.

 

BRIT 3D color model
Since we have both the single high-resolution image and 3D data about the landscape, we are able to merge the two together in a 3D color model that lets us visualize the landscape from any angle.

 

Next Steps

We will continue to survey our BRIT prairie and record changes over time, but we also intend to expand our aerial survey program to include surveys of other, larger landscapes and ecosystems as well as to experiment with the use of unmanned aerial vehicles (UAVs), also known as remote controlled planes and copters. Keep your eyes open for more posts about some of the technologies behind the image analysis and future aerial surveys!

Leave A Response

Plain text

  • No HTML tags allowed.
  • Lines and paragraphs break automatically.
  • Web page addresses and email addresses turn into links automatically.

Related Articles

Aerial Survey Resources – For Beginners

This page is constantly evolving as a resource for conducting aerial surveys. At this point, it’s not meant to be comprehensive or provide everything you need to get started. It’s just to give you a general framework for the process and to point you to other resources. If you have questions or suggestions, please comment below. The blog post describing BRIT’s initial aerial surveys is at http://brit.org/phytophilia/aerial-surveys-at-BRIT and provides a bit more background about some of the reasons BRIT is conducting these surveys. Starting a new aerial survey project generally involves the following steps: Purchase/gather/make survey kit Perform survey Process images Survey Kit One of the best places to get started is at Public Lab ( http://publiclab.org ). This group provides a wealth of...
Read More >

Mary Sophie on the Move: Busted Freezer Gets Relocated, Repurposed

By Research Associate Will Godwin, PhD Adaptive reuse or re-purposing has become a popular method to achieve green or sustainable design. It even extends into the aesthetics of interior design through shabby-chic and the more avant-garde, lab-chic. But reuse is not a recent idea. It has a long history in the natural world. Charles Darwin was the first to call it “pre-adaptation.” By the 1980’s the term was adjusted to “exaptation” primarily to avoid connotations of premeditation. Exaptation happens when a structure or behavior that was useful for one function is suddenly and serendipitously useful in a new way. Animals and plants have been finding fortuitous new uses for stuff they already had for millions of years. Often it gave the survivors that unexpected edge required to survive. The...
Read More >

A 54-Year Celebration

The Journal of the Botanical Research Institute of Texas ( JBRIT ) is celebrating its 54th year of continuous publication. It all started when Lloyd H. Shinners —a member of the Southern Methodist University (SMU) faculty and a prolific botanical researcher and writer who wanted to edit his work and the work of others—founded and published the first two issues of Sida, Contributions to Botany on November 23, 1962. He named the journal for a genus ( Sida ) of yellow-flowered plants of the mallow or cotton family (Malvaceae), distributed throughout the world and especially common in Texas. Shinners served as editor and publisher until his death in 1971, after which William F. Mahler, professor of botany at SMU, inherited the journal and continued to edit and privately publish it. Barney Lips...
Read More >

The Living Herbarium: Instructions for Life

Article originally published in The Leaflet (April 2014) by Brian Witte, PhD, BRIT Research Associate (Disclaimer: The technical aspects of this article are dramatically simplified in the interests of communicating with an audience entirely unfamiliar with molecular biology. Send me an email ( bwitte@brit.org ) if you would like a deeper explanation.) We like to repeat, loudly and often, that there are over 1 million plant specimens in the Philecology Herbarium at BRIT. It’s a nice, big, round number, and it sounds cool when tour groups come through. What if I told you that as imposing as that number sounds, the real number is closer to a thousand billion (1,000,000,000,000) plants*? The goal of a herbarium is to preserve plants. The ideal specimen, in many respects, has all the essential...
Read More >