Researchers from Princeton University and the University of Washington have co-developed a micro camera sensor that is as small as a grain of salt and can be used in tiny robots for endoscopy and brain imaging more effectively than existing cameras. The researchers said the sensor is just 500microns (one-millionth of a meter) in diameter and is built with a computationally designed metasurface that includes 1.6 million cylindrical posts called nanoantennas. Using machine learning algorithms the antennas combine their interactions with light to generate high-quality images.
The nanoantennas are placed inside silicon nitride, a glass-like material that is commonly used in semiconductor manufacturing.
Though tiny cameras have been developed before, they suffer from image distortion, limited field of view and the inability to capture the full spectrum of colours. Typically cameras use several curved glass or plastic as lenses to capture high-quality images and that restricts attempts to miniaturise them.
Researchers claim the new camera can generate images that look as crisp and vivid as those captured by existing compound camera lenses that are 500,000 times larger in size.
“Ultrathin meta-optics utilise subwavelength nano-antennas to modulate incident light with greater design freedom and space-bandwidth product over conventional diffractive optical elements,” the researchers wrote in the paper that was published in the journal Nature Communications.
What sets the new camera sensor apart from any other existing cameras is the integrated design of the optical surface and the signal processing algorithms that produce the image, points out Felix Heide, senior author and assistant professor, computer science at Princeton University.
“This boosted the camera’s performance in natural light conditions, in contrast to previous metasurface cameras that required the pure laser light of a laboratory or other ideal conditions to produce high-quality images,” said Heide in a separate post on Princeton University website.
Designing and configuring the nanoantennas to get desired results was not easy, as per the researchers. They, however, addressed it by creating a computational simulator to automate the testing of different nano-antenna configurations. “Because of the number of antennas and the complexity of their interactions with light, this type of simulation can use massive amounts of memory and time,” rued Shane Colburn, affiliate assistant professor in the Department of Electrical and Computer Engineering at the University of Washington. Colburn developed a model that could efficiently and accurately approximate the metasurfaces’ image production capacity.
The researchers believe that further miniaturisation can open new avenues for the use of cameras, especially in the field of nanorobotics, augmented reality (AR) and virtual reality (VR) and more areas of healthcare monitoring. They also said that the metasurface design can be produced at scale and at a lower cost than the lenses used in conventional cameras.