EVS and Argonne researchers developing an AI solution for monitoring birds at solar facilities
As solar energy expands across the United States, scientists are only just beginning to quantify the effects on wildlife. The U.S. Department of Energy's (DOE) Argonne National Laboratory has been awarded $1.3 million from DOE's Solar Technologies Office to develop technology that can cost-effectively monitor avian interactions with solar energy infrastructure.
The three-year project, which began in March 2020, will combine computer vision techniques with a form of artificial intelligence called deep learning to monitor solar energy sites for birds and collect data about what happens when they approach solar panels.
The new project aims to reduce the frequency of human surveillance by using cameras and computer models that can collect more and better data at a lower cost. Achieving that involves three tasks: detecting moving objects near solar panels; identifying which of those objects are birds; and classifying events (such as perching, flying through, or a collision). Using deep learning, scientists build models using methods that will "teach" computers how to spot birds and behaviors by feeding in examples. Once the model is trained, it will run internally within the cameras on a live video feed, classifying interactions on the fly.
Read the full article by Christina Nunez.
Since the early-stage capability development, Argonne’s Sustainability Program has been supporting the project by providing access to the solar facility on the Argonne site for video collection, instrument testing, and simulation of bird activities. The Avian-Solar DL project aligns with Argonne’s effort to promote environmental sustainability.
Argonne’s Computer, Environment, Life Sciences Directorate supports video annotation, a critical task for generating a large, quality training dataset for developing accurate and robust AI for bird monitoring.