As the EMVA Conference in Rome drew toward its conclusion, one of the final speakers managed to inject a fresh burst of energy into the room. Fabrizio Preda, CEO and co-founder of NIREOS, stepped onto the stage with a mission: to convince the machine vision world that hyperspectral imaging isn’t just a niche—it’s the future of precise, data-rich sensing.
Founded in 2018 as a university spin-off, the Milan-based company is now a leader in compact, high-performance spectroscopy devices. NIREOS has matured rapidly over its seven years, thanks in part to early support from European research projects and collaborations with the European Space Agency and major international companies.
Beyond the RGB: The Power of Hyperspectral
For the uninitiated, Preda provided a primer: unlike traditional cameras that capture images in three color channels—red, green, and blue—hyperspectral imaging collects dozens to hundreds of spectral bands at each pixel. Think of it as building a 3D cube of data, where every pixel contains a full light spectrum, from ultraviolet to infrared.
While powerful, capturing this data isn’t trivial. Most existing systems rely on snapshot arrays, filter wheels, or “pushbroom” scanning—each with compromises in speed, resolution, or complexity.
NIREOS, however, has charted its own path.
Reinventing the Interferometer
Rather than relying on conventional approaches, NIREOS employs a method based on Fourier Transform Spectroscopy (FTS)—an established concept in physics but rarely seen in compact imaging devices.
At the core of NIREOS’s system is a novel interferometer that manipulates light delays with high precision using birefringent crystals. This breakthrough allows the device to function not only in the infrared (where FTS traditionally operates), but also in the visible and ultraviolet—a longstanding challenge in the field.
The result? A passive, robust module called GEMINI that transforms nearly any monochrome camera into a hyperspectral imaging system. It’s placed between the camera lens and sensor, with no moving parts during image acquisition. Preda emphasized its versatility: “It works with CCDs, sCMOS, microbolometers—you name it. And you can use it for anything from microscopes to drones.”
Imaging What Others Can’t
Hyperspectral imaging’s real power lies in its applications, and Preda’s talk was filled with vivid examples:
- In life sciences, NIREOS systems help researchers distinguish subtle differences in autofluorescence from mouse lung tissues—without dyes. The technology allows for the detection of fibrotic changes invisible to the human eye or standard imaging.
- In medical diagnostics, they’re developing a system with a hospital in Pavia to detect breast cancer tissue margins in the operating room—cutting diagnosis time from days to seconds. “The tumor was invisible in RGB. At 1600 nm, it popped out,” Preda said.
- In renewable energy, the system detects microscopic defects in silicon and perovskite solar cells via electroluminescence. This information is critical for improving efficiency in the next generation of photovoltaic technology.
- In vertical farming, NIREOS cameras monitor crops for early signs of disease and optimize nutrient recipes by analyzing reflectance and fluorescence signals. In one project, the team used hyperspectral imaging to track chlorophyll decay and shelf-life in harvested greens.
When Less Is More
But Preda also offered a note of pragmatism. Hyperspectral data is powerful—but not always necessary. Once the key wavelengths are identified in research using hyperspectral tools, companies can deploy cheaper, faster multispectral systems tailored to their specific needs.
“It’s about knowing where to look,” Preda noted. “We provide the R&D capability, but also support companies in scaling to in-line, real-time solutions.”
A Vision Beyond Vision
As his presentation drew to a close, Preda listed other domains where NIREOS is active: food inspection, cultural heritage preservation, microplastics detection, microbiology, and more. The unifying thread? Imaging the invisible.
In a field crowded with hype, Preda’s approach stood out—grounded, rigorous, and refreshingly clear-eyed about the technology’s strengths and limits.
And in doing so, he made a convincing case: hyperspectral imaging isn’t just about seeing more. It’s about seeing what truly matters.
For more coverage of the EMVA conference 2025 in Rome, click HERE
