Princeton and University of Washington researchers have built a grain-sized camera capable of capturing clear, full-color images. It’s composed of a light-capturing metasurface that could eventually be scaled up to transform entire surfaces into sensors.
Existing micro cameras use metasurfaces, but physics limits their image quality. A typical camera creates sharp images by bending light into focus through a series of lenses. However, the camera must be large enough to accommodate those lenses.
“Meta-optics,” which utilizes exceedingly small nanostructures that can capture and re-emit light, is one way to shrink a camera. Each of these miniature structures collects the photons that form a light, and the metasurface then converts those photons into a signal that a computer can comprehend. Consider them to be tiny antennas.
The new device resembles a transparent panel with an engraved circular pattern. Each of the 1.6 million cylinders in that half-millimeter-wide circle is precisely constructed to bend light in the appropriate direction so that the array bends the optical wavefront. After that, signal processing algorithms create an image from the data.
The images produced are significantly more precise than those of other tiny sensors. In experiments, the team researchers demonstrated that the new sensor acquired images with a spatial resolution of 214-line pairs per millimeter and a resolution of 720 x 720 pixels in actual color, catching wavelengths between 400 and 700 nm in natural light. In addition, the device has a field of view of 40 degrees and an f-number of 2. The team claims the images are comparable to those obtained by a traditional compound camera lens half a million times larger than their new sensor.
The post-processing algorithms handle a lot of the hard lifting regarding image quality improvement. They were created in tandem with the metasurface to ensure they functioned well together. Another benefit is that the devices are simple to manufacture, implying that they should be relatively easy to scale up for mass production. They are composed of silicon nitride, and the surface nanostructures can be made with deep ultraviolet lithography, a process already being used to generate semiconductors.
According to the researchers, these devices could be used for medical imaging in the body, offering an improved vision to small robots, or even turning almost anything into a camera by lining its surface with thousands of these sensors.
Felix Heide, the senior author of the study, said:
“We could turn individual surfaces into cameras that have an ultra-high resolution so that you wouldn’t need three cameras on the back of your phone anymore, but the whole back of your phone would become one giant camera. So we can think of completely different ways to build devices in the future.”
The research was published on November 29, 2021, in Nature Communications.
