One of the hardware fields that is receiving an exceptional amount of research and development today is imaging systems. Between pursuing better smartphone cameras or advanced imaging systems for autonomous vehicles, imaging hardware is a dynamic and vital technology. 

Recently, researchers at NIST announced their research into a new imaging lens based on metalenses that claim to achieve a record-setting depth of field. Notably, the new technology was inspired by an unlikely source: prehistoric trilobites.

NIST researchers claim to have been inspired by ancient trilobites when creating their latest imaging technology. Image used courtesy of NIST

In this article, we’ll discuss metalenses and take a closer look at the NIST research to understand the technology and its potential impact on the field.

Metalenses for Better Image Sensors and Devices

Within optical engineering, one of the most exciting developing technologies is metalenses.

Metalenses are a novel lens technology that consists of arrays of nanostructures known as “nanoantennas” placed on the surface of a specially crafted flat lens. 

Within a metalens, the nanoantennas are designed in such a way that they can selectively induce the phase of incident light. In simpler terms, the nanoantennas allow metalenses to manipulate and focus light in specific ways that would have otherwise required conventional refractive lenses.

Metalenses (C) use nanoantennas for phase control unlike conventional refractive lenses (B). Image used courtesy of Engleberg et al

There are many advantages to metalenses over conventional diffractive lenses (CDLs). Some crucial benefits include a reduced thickness compared to CDLs and tunability: the ability to change focus and zoom adjustments of the lens dynamically. Beyond this, metalenses have more degrees of freedom than CDLs since their nanostructures come in unlimited forms. 

Thanks to these advantages, metalenses have been praised as a potential path forward for the miniaturization of optical sensors and different imaging devices.

A Nature-inspired Camera

This week, researchers at NIST made headlines with their recent paper introducing a new, "record-setting" lens for imaging solutions.

As described, the new lens takes inspiration from the prehistoric trilobite, a creature who boasted incredible vision thanks to compound eyes: eyes that are composed of thousands of independent units as opposed to a single, larger unit. 

In their research, the NIST scientists sought to mimic the vision systems of a specific type of trilobite whose eyes were composed of two lenses that bent light at different angles, allowing the creature to see with equal clarity at near distances and extremely far ranges.

The new metalens from NIST claims to take sharp images at a range of centimeters or kilometers. Image used courtesy of S. Kelley/NIST

To do this, the researchers created a miniature camera based on a bifocal metalens array consisting of millions of nanoantennas etched into the flat glass. These nanoantennas were shaped and oriented to focus light on nearby objects equally and look for objects at a longer range. 

The researchers then coupled this new imaging hardware with a specially designed multiscale convolutional neural network (CNN) to correct for optical aberrations caused by the metalens. The CNN also served to sharpen any received images that occurred at distances between the focal lengths of the bifocal camera.

NIST's camera system (a) and working principle (b). Image used courtesy of Fan et al

The result of this combination of nanophotonic technology and computational photography was a camera that was able to take high-quality images that were equally as sharp at a distance of 3 cm as at a distance of 1.7 km. 

According to the researchers, their results represent a record-setting depth of field (i.e., the distance over which a camera can take a sharp image).

A New Direction for Imaging Hardware

With this new research, the NIST scientists have done multiple important things. 

First, with their new camera, the researchers have introduced new, impressive hardware that could have important repercussions for fields like computer vision. Beyond this, their work is vital in that it shows the promising results that are possible at the intersection of machine learning and nanophotonics. 

As imaging hardware continues to be developed and improved, this new research from NIST will hopefully inspire a new path and direction for the future of the field.