Camera see the world differently from humans. Resolution, equipment, lighting, distance, and atmospheric conditions can affect how a person interprets objects in a photo. For Sophie Voisin, a software engineer at the Department of Energy’s Oak Ridge National Laboratory, imaging can reveal what the human eye can’t see, offering a different perspective to understanding the changing world. how everyday.
Technically, Voisin’s work at ORNL is all about imaging – enhancing, improving, analyzing and exploiting high-altitude and low-altitude images and full-motion video from drones. Every day, satellites use cameras, called sensors, to see what’s happening around the world. The volume of images taken every day is huge. This is where Voisin’s true passion for engineering comes into play.
Sophie Voisin, an ORNL software engineer, is part of the team that won the 2014 R&D 100 Award for work on Intelligent Software for Personal Modeling for Expert Opinion, Decisions and Errors in Visual Inspection Tasks. Image supplier: Jason Richards/ORNL, US Department of Energy
“I love coding. I like the technical side of programming algorithms to find the hot spot to apply the latest and greatest research to real-life problems,” says Voisin. “I can immediately see if changes to my algorithm are being applied properly.”
Maintain speed, accuracy
Voisin has nurtured algorithms over the past seven years to process ever-growing collections of images faster. What started out as a small app has grown to 12 projects; Now she leads a team of 32 people as lead investigator. The team aims to use software development, scientific research, and software engineering to figure out which images show changes to the landscape that may be of interest to the US government. By allowing the computer to sort through the initial blocks of images and flag certain pictures, analysts can then review and interpret which images are truly valuable to the audience. with decision makers.
Image processing is not an easy process. The number of raw images continues to grow as sensors take more and more snapshots of the world. Processing each image will increase its size by two or three times. With large file sizes that make up huge data sets, high-performance computing offers the best chance of processing images at the speed needed to make the right decisions. Leveraging the speed of ORNL’s high-performance computers, Voisin’s team also strives to provide a high degree of confidence in the accuracy of the results.
Over the past half-decade, Voisin has adapted and maintained applications as hardware changed and graphics cards increased in capabilities. To prevent the algorithm from breaking, Voisin and her team test new hardware before it goes live to ensure a constant flow of information.
Algorithms make the difference
While images are images, Voisin says, it’s the size of the different files. Since embarking on a doctoral program in computer science and image processing, she has worked on beams at the High Flux Isotope Reactor, a facility used by the DOE Office of Science, about medical imaging for mammography and now geospatial imaging. For her, the challenge of developing and adapting algorithms for different types of images is the fun part.
“National security work is a different pace than other industries. I can test algorithms to fit the data and then upload it to the project,” says Voisin. But to apply machine learning algorithms to data, computer scientists need training data.
The training data helps the model learn what to look for in an image. A new algorithm may not know the difference between a car and a house, since both can appear as squares. Computer scientists write algorithms to detect certain features and then train the algorithm to distinguish between “squares” and determine which ones need to be flagged for further analysis.
Continue to learn from colleagues
East Tennessee has been home to Voisin and her family since moving to the United States from France more than a decade ago. She and her husband found opportunities at ORNL to work on compelling and impactful projects with applied research.
“I go to the lab for a research opportunity, but the day-to-day work is about the people you work with and I am so grateful to have worked with such a great team,” says Voisin. “Having a good relationship with my colleagues pushes me to be a good mentor because I still learn a lot from others.”
Source: ORNL
