A recent article in the journal Sensors highlights the Galileo Project, an initiative tackling gaps in the scientific study of unidentified aerial phenomena (UAP).
Study: Commissioning an All-Sky Infrared Camera Array for Detection of Airborne Objects. Image Credit: MR.PRAWET THADTHIAM/Shutterstock.com
The project is focused on creating an advanced ground-based observatory that continuously monitors the sky using a multi-modal, multi-spectral approach. By running a long-term aerial survey, researchers aim to collect extensive data on UAP, helping to build a clearer understanding of these phenomena.
Background
Concerns over the lack of high-quality UAP data have been echoed by US federal agencies, including the Office of the Director of National Intelligence (ODNI). ODNI's annual reports have repeatedly called for better data collection, while a 2023 NASA study emphasized the need for precision-calibrated sensors and AI-driven detection methods.
The Galileo Project is addressing these concerns by deploying multiple sensor types over extended periods, allowing researchers to track seasonal variations and long-term trends—an approach that sets it apart from previous studies.
The Current Study
At its Massachusetts observatory, the Galileo Project has assembled a sophisticated array of instruments. The focal point is an all-sky infrared camera system featuring eight uncooled long-wave infrared FLIR Boson 640 cameras. These cameras use both intrinsic and extrinsic calibration techniques, incorporating real-time Automatic Dependent Surveillance–Broadcast (ADS-B) data to improve tracking accuracy.
To analyze aerial activity, the project relies on state-of-the-art computer vision technology. A machine learning model called You Only Look Once (YOLO) identifies objects, while the Simple Online and Realtime Tracking (SORT) algorithm reconstructs trajectories. Over five months, the system recorded approximately 500,000 aerial trajectories, with detection efficiency influenced by weather conditions, detection range, and aircraft size.
Researchers are using an outlier detection approach to flag unusual trajectories. So far, about 16 % of the data has been identified as potential anomalies, triggering manual review for further analysis.
Results and Discussion
Initial findings from the Galileo Project indicate a 41 % detection acceptance rate for ADS-B-equipped aircraft, with a frame-by-frame detection efficiency of 36 %. The system's overall aircraft reconstruction rate stands at 13.4 %, forming a foundation for future refinements in tracking and detection technology.
Among the 80,000 flagged outliers, 144 trajectories remain ambiguous. While most likely represent ordinary objects, the lack of detailed data—such as precise distance and kinematics—hinders further classification. Moving forward, the project will focus on validating its findings and improving classification methods.
Beyond detection, the Galileo Project aims to statistically analyze UAP patterns over time and across various regions. Plans are in place to expand the observatory network, enhance data collection, and reinforce the scientific rigor of UAP research.
Conclusion
Positioned at the forefront of UAP research, the Galileo Project is leveraging advanced sensor technology and AI-driven data analysis to address longstanding knowledge gaps. By refining detection methods and expanding its observational reach, the project is setting a new standard for aerospace anomaly research.
As investigations continue, the findings could reshape our understanding of aerial phenomena, challenge existing theories, and inspire deeper discussions within the aerospace community.
Journal Reference
Domine L., Biswas A., et al. (2025). Commissioning an All-Sky Infrared Camera Array for Detection of Airborne Objects. Sensors 25(3):783. DOI: 10.3390/s25030783, https://www.mdpi.com/1424-8220/25/3/783