Researchers at AMOLF (Amsterdam, Netherlands) and their collaborators from the Advanced Science Research Center (ASRC/CUNY; New York, NY) have created a nanostructured surface capable of performing on-the-fly mathematical edge-detection operations on an input image.1 This discovery could boost the speed of image processing and lower energy usage. The work enables ultrafast object detection and augmented-reality applications.
Image processing is at the core of multiple rapidly growing technologies, such as augmented reality, autonomous driving, and more general object recognition -- and edge detection in an image is what typically becomes the starting point for image recognition. Edge detection is typically performed either via computer algorithms, which implies fundamental speed limitations and high energy consumption; or in an analog fashion, which requires bulky optics.
Nanostructured metasurface
In a completely new approach, AMOLF PhD student Andrea Cordaro and his coworkers created an optical metasurface in the form of a transparent substrate holding an array of silicon nanobars. When an image is projected onto the metasurface, the transmitted light forms a new image that shows the edges of the original. Effectively, the metasurface performs a mathematical derivative operation on the image, which provides a direct probe of edges in the image.
In a first experiment, an image of the AMOLF logo was projected onto the metasurface. At the metasurface's resonance wavelength (726 nm), a clear image of the edges is observed. The mathematical transformation results from the fact that each spatial frequency that composes the image has a tailored transmission coefficient through the metasurface. This tailored transmission is the result of a complex interference of light as it propagates through the metasurface.
Edge detection
To demonstrate edge detection experimentally on an image, the researchers, created a miniature version of the painting Meisje met de parel (by Vermeer) by printing chromium dots onto a transparent substrate. If the image is projected onto the metasurface using off-resonant illumination (750 nm), the original image is clearly recognized. In contrast, if the illumination has the right color (726 nm), the edges are clearly resolved in the transformed image.
The new optical computing and imaging technique operates at the speed of light and the mathematical operation itself consumes no energy, as it involves only passive optical components. The metasurface can be readily implemented by placing it directly onto a standard CCD or CMOS detector chip, opening new opportunities in hybrid optical and electronic computing that operate at low cost, low power, and small dimensions.
Source: https://amolf.nl/news/mathematics-at-the-speed-of-light
REFERENCE:
1. A. Cordaro et al., Nano Letters (2019) doi.org/10.1021/acs.nanolett.9b02477.
Got optics- and photonics-related news to share with us? Contact John Wallace, Senior Editor, Laser Focus World
Get more like this delivered right to your inbox
