Working principle of RF photonic video image processor. PD photodetector. a diagram illustrating the flattening method applied to the input video frames, both horizontally and vertically. b Schematic illustration of an experimental setup for video image processing. c The video images processed after (i) 0.5 order differentiation for edge detection, (ii) integration for motion blue, and (iii) Hilbert transform for edge enhancement. Credit: Communications Engineering (2023). DOI: 10.1038/s44172-023-00135-7
A team of international scientists has developed an ultra-fast signal processor capable of simultaneously analyzing 400,000 video frames in real time, according to a paper published in Communications Engineering.
The team, led by Professor David Moss from Swinburne University of Technology, developed a processor that operates more than 10,000 times faster than typical electronic processors operating at gigabytes/s, at a record rate of 17 terabits /s (trillion bits per second).
This technology has profound implications for the safety and efficiency of driverless cars, medical imaging, and could help find habitable planets beyond our solar system.
“It’s a game changer,” Moss says. “These are portable, ultra-precise optical rulers that could be used for everything from portable respiratory scanners to detecting small movements of objects in space or searching for Earth-like exoplanets.”
CPU efficiency and speed will have a major impact on artificial intelligence (AI), machine learning and robotic vision. The ability to process large amounts of data in real time will lead to more intelligent and autonomous robotic systems capable of operating effectively in real-world environments.
In astronomy, for example, processing large volumes of celestial data at unprecedented speeds will enable revolutionary discoveries. This approach will also accelerate medical imaging and autonomous vehicle functions.
“All real-time analysis and decision-making requires ultra-fast analysis of video images, from autonomous driving to artificial intelligence, to detect and analyze objects.”
“The secret to its remarkable performance lies in the integrated optical microcomb, which overcomes the bandwidth and energy limitations inherent in electronics,” says Moss.
Moss says this photonic signal processor is not just a technological marvel: it’s a harbinger of a future where efficient and rapid data processing will shape every aspect of our lives.
“From transforming transportation and healthcare to opening new doors in space exploration and AI, the impact of this technology will be profound and far-reaching.”
More information:
Mengxi Tan et al, Kerr microcomb-based photonic signal processor for real-time video image processing, Communications Engineering (2023). DOI: 10.1038/s44172-023-00135-7
Provided by Swinburne University of Technology
Quote: New ultra-fast processor to advance AI, driverless vehicles and more (December 21, 2023) retrieved December 21, 2023 from
This document is subject to copyright. Apart from fair use for private study or research purposes, no part may be reproduced without written permission. The content is provided for information only.
