Even NASA knows that it can't always rely on satellite imaging when a natural disaster strikes. This is why Igor Carron, Assistant Director of the Spacecraft Technology Center at Texas A&M University (TAMU), recently used a stratospheric balloon with several of his students. They've used a simple point-and-shoot digital camera to record hundreds of images over New Mexico. And by using inexpensive commercial software, they've stitched together these images to create large panoramas of up to 150 km which are as accurate as the more expensive maps produced by NASA or companies such as DigitalGlobe which sells data to Google. Igor predicts that "traditional GIS will be replaced by user fed data and applications" and that his project is just the beginning of "remote sensing for the people by the people." Here are some excerpts of a conversation I had with him.
Let's start with a short explanation of the GeoCam project extracted from this overview.
High altitude balloons can be deployed in a matter of hours and provide emergency remote sensing thereby enabling first responder's situational awareness and give adequate trajectories to rescuers. It would also enable the diffusion of imagery to a large segment of the population looking for information thereby reducing the strain on the telecommunication system of the affected area.
And on September 4, 2006, Igor Carron, Assistant Director of the Spacecraft Technology Center at Texas A&M University (TAMU), and his students took place as a 'payload application' of the Louisiana State University's High Altitude Student Payload (HASP), sponsored by NASA's Balloon Program Office. And they took more than 4 GB of pictures. Here is a comment from Igor about how they originally wanted to use these pictures.
Originally their intent was to integrate one by one all these images on an application like Google Maps or Google Earth using the GPS information provided from the balloon. They soon figured out that this was too complex and would not be imitated in the real world with people who would not have the combined skills of photographic overlays integration using the Google Maps APIs, orthographics projections and Adobe's PhotoShop.
But when they started to use PhotoShop, they soon realized that they needed several dozens of gigabytes for working space. So they decided to change strategy to produce their large panoramas.
Below are two of these panoramic images. The one on the top gathers the images shot between 4 and 5 PM on September 4, 2006 while the one the bottom illustrates the path of the balloon during the next hour (Credit: Igor Carron and his team). Here are two links to larger versions of these panoramas, here and there.
And below is the full path for the flight during which 609 photos were shot (Credit: Igor Carron and his team). Here is a link to a larger version.
For those of you who want to know the details, these panoramic images have been stitched together with Autopano Pro, a commercial implementation of the AutoStitch software that I've enthusiastically reviewed in April 2005 (Check "Build Panoramic Photographies with AutoStitch"). Then the images, which were 'weighing' dozens of gigabytes, have been put online by using another commercial software, Zoomify.
Now, here is a first question to Igor: What was the purpose of this survey over New Mexico?
It was a student experiment. We were just one of the payload of the HASP facility designed by LSU (Louisiana State University). The balloon is a NASA balloon that is coming from the NASA's Columbia Scientific Balloon Facility located in Palestine, Texas. NASA funds this activity to get students involved in near space environment and the engineering associated with it. So NASA is involved has a funding agency for the balloon and is also providing staff to launch that balloon. I believe they are also helping LSU define what kind of payloads can or cannot fly out of all the student's team submissions. The reason it flew over New Mexico has to do with the fact that sometimes NASA launches their balloons from Fort Sumner, NM.
And why using images from balloons and not from satellites?
In this experiment, our team proposed a subject i.e. flying the latest cheap off the shelf camera to fly on that platform and evaluate its potential for remote sensing. LSU allowed us to be one of the experiments to fly.
The big difference with satellites is really the following. With satellites, you are bound to follow its orbit and you cannot get an assessment of the ground conditions until the satellite flies over the area (which can take up to three days). And then, when the satellite flies over the area of interest, there should not be any clouds otherwise you are bound to wait for another three days. In our case, we have the same issue with cloud cover but we can also launch a balloon or use a UAV in a much more responsive manner.
In case of disasters of enormous proportion such as Hurricane Katrina, Igor thinks that armed with cheap cameras, inexpensive balloons and easy-to-use software, almost everyone could provide quality maps to first time responders.
Finally, if you're interested in this project, here is a link to its blog.
Sources: Igor Carron, Spacecraft Technology Center, Texas A&M University (TAMU), December 2006; and various websites
You'll find related stories by following the links below.