New AI Sensor 'Sniffs' Out Spectral Targets

Berkeley Lab researchers develop an intelligent sensor that can identify chemicals and characterize materials quickly and efficiently.

Published on Feb. 12, 2026

Got story updates? Submit your updates here. ›

Researchers at the Department of Energy's Lawrence Berkeley National Laboratory have developed an AI-enhanced spectral imaging sensor that can "sniff and seek" target objects in real-time. By integrating machine learning algorithms directly into the sensor hardware, the device can perform object identification and material characterization tasks much faster and more efficiently than traditional spectral imaging systems.

Why it matters

Spectral imaging tools are vital for applications like semiconductor fabrication, pollutant tracking, and crop monitoring, but have traditionally been limited by slow processing speeds and high power consumption. This new sensor design addresses those bottlenecks, paving the way for more widespread adoption of spectral machine vision technologies.

The details

The sensor design integrates the AI computation and spectral analysis directly into the image capture process, rather than relying on separate sensor and computational modules. This allows the sensor to "sniff" and identify target objects much faster and more efficiently than traditional systems. The researchers trained the sensor to identify birds in images by showing it examples of bird plumage versus background, and the sensor was then able to successfully identify birds in new images it had never seen before.

  • The research was published in a study in February 2026.

The players

Ali Javey

A senior faculty scientist at Berkeley Lab and a professor of materials science and engineering at UC Berkeley, who led the research study.

Dehui Zhang

A postdoc in Berkeley Lab's Materials Sciences Division and the lead author on the study.

Aydogan Ozcan

A researcher at UCLA who collaborated closely with the Berkeley Lab team on the project.

Department of Energy's Lawrence Berkeley National Laboratory

The research institution where the sensor was developed.

Got photos? Submit your photos here. ›

What they’re saying

“We focused on enhancing the speed, resolution, and power efficiency of existing spectral machine vision technologies by more than two orders of magnitude.”

— Ali Javey, Senior faculty scientist at Berkeley Lab (Mirage News)

“Photodetection can be perceived as an automatic physical computational process.”

— Dehui Zhang, Postdoc in Berkeley Lab's Materials Sciences Division (Mirage News)

“For me, the most exciting part is the concept of giving intelligence to sensors.”

— Ali Javey, Senior faculty scientist at Berkeley Lab (Mirage News)

What’s next

The researchers plan to continue exploring applications for the intelligent sensor technology beyond spectral imaging, including in other advanced optical sensing fields.

The takeaway

This new sensor design represents a significant advancement in spectral machine vision, addressing longstanding challenges around processing speed and power efficiency. By integrating AI directly into the sensor hardware, the technology opens the door for more widespread adoption of spectral imaging tools in a variety of real-world applications.