VCSEL-based photonic spiking neural networks for ultrafast detection and tracking
Inspired by efficient biological spike-based neural networks, we demonstrate for the first time the detection and tracking of target patterns in image and video inputs at high-speed rates with networks of multiple artificial spiking optical neurons. Using photonic systems of in-parallel spiking vert...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IOP Publishing
2024-01-01
|
| Series: | Neuromorphic Computing and Engineering |
| Subjects: | |
| Online Access: | https://doi.org/10.1088/2634-4386/ad2d5c |
| _version_ | 1827318918623002624 |
|---|---|
| author | Joshua Robertson Paul Kirkland Gaetano Di Caterina Antonio Hurtado |
| author_facet | Joshua Robertson Paul Kirkland Gaetano Di Caterina Antonio Hurtado |
| author_sort | Joshua Robertson |
| collection | DOAJ |
| description | Inspired by efficient biological spike-based neural networks, we demonstrate for the first time the detection and tracking of target patterns in image and video inputs at high-speed rates with networks of multiple artificial spiking optical neurons. Using photonic systems of in-parallel spiking vertical cavity surface emitting lasers (VCSELs), we demonstrate the implementation of multiple convolutional kernel operators which, in combination with optical spike signalling, enable the detection and tracking of target features in images/video feeds at an ultrafast photonic operation speed of 1 ns per pixel. Alongside a single layer optical spiking neural network (SNN) demonstration, a multi-layer network of photonic (GHz-rate) spike-firing neurons is reported where the photonic system successfully tracks a large complex feature (Handwritten Digit 3). The consecutive photonic layers perform spike-enabled image reduction and convolution operations, and interact with a software-implemented SNN, that learns the feature patterns that best identify the target to provide a high detection efficiency even in the presence of a distractor feature. This work therefore highlights the effectiveness of combining neuromorphic photonic hardware and software SNNs, for efficient learning and ultrafast operation, thanks to the use of spiking light signals, towards tackling complex AI and computer vision problems. |
| first_indexed | 2024-04-25T00:10:57Z |
| format | Article |
| id | doaj.art-2485a2ad71f14b69838d1a6c6674e765 |
| institution | Directory Open Access Journal |
| issn | 2634-4386 |
| language | English |
| last_indexed | 2024-04-25T00:10:57Z |
| publishDate | 2024-01-01 |
| publisher | IOP Publishing |
| record_format | Article |
| series | Neuromorphic Computing and Engineering |
| spelling | doaj.art-2485a2ad71f14b69838d1a6c6674e7652024-03-13T12:28:32ZengIOP PublishingNeuromorphic Computing and Engineering2634-43862024-01-014101401010.1088/2634-4386/ad2d5cVCSEL-based photonic spiking neural networks for ultrafast detection and trackingJoshua Robertson0https://orcid.org/0000-0001-6316-5265Paul Kirkland1Gaetano Di Caterina2Antonio Hurtado3Institute of Photonics, SUPA Department of Physics, University of Strathclyde , Glasgow, United KingdomDepartment of Electronic and Electrical Engineering, University of Strathclyde , Glasgow, United KingdomDepartment of Electronic and Electrical Engineering, University of Strathclyde , Glasgow, United KingdomInstitute of Photonics, SUPA Department of Physics, University of Strathclyde , Glasgow, United KingdomInspired by efficient biological spike-based neural networks, we demonstrate for the first time the detection and tracking of target patterns in image and video inputs at high-speed rates with networks of multiple artificial spiking optical neurons. Using photonic systems of in-parallel spiking vertical cavity surface emitting lasers (VCSELs), we demonstrate the implementation of multiple convolutional kernel operators which, in combination with optical spike signalling, enable the detection and tracking of target features in images/video feeds at an ultrafast photonic operation speed of 1 ns per pixel. Alongside a single layer optical spiking neural network (SNN) demonstration, a multi-layer network of photonic (GHz-rate) spike-firing neurons is reported where the photonic system successfully tracks a large complex feature (Handwritten Digit 3). The consecutive photonic layers perform spike-enabled image reduction and convolution operations, and interact with a software-implemented SNN, that learns the feature patterns that best identify the target to provide a high detection efficiency even in the presence of a distractor feature. This work therefore highlights the effectiveness of combining neuromorphic photonic hardware and software SNNs, for efficient learning and ultrafast operation, thanks to the use of spiking light signals, towards tackling complex AI and computer vision problems.https://doi.org/10.1088/2634-4386/ad2d5cneuromorphic computingsemiconductor laserspiking neural networksimage processingdetection and tracking |
| spellingShingle | Joshua Robertson Paul Kirkland Gaetano Di Caterina Antonio Hurtado VCSEL-based photonic spiking neural networks for ultrafast detection and tracking Neuromorphic Computing and Engineering neuromorphic computing semiconductor laser spiking neural networks image processing detection and tracking |
| title | VCSEL-based photonic spiking neural networks for ultrafast detection and tracking |
| title_full | VCSEL-based photonic spiking neural networks for ultrafast detection and tracking |
| title_fullStr | VCSEL-based photonic spiking neural networks for ultrafast detection and tracking |
| title_full_unstemmed | VCSEL-based photonic spiking neural networks for ultrafast detection and tracking |
| title_short | VCSEL-based photonic spiking neural networks for ultrafast detection and tracking |
| title_sort | vcsel based photonic spiking neural networks for ultrafast detection and tracking |
| topic | neuromorphic computing semiconductor laser spiking neural networks image processing detection and tracking |
| url | https://doi.org/10.1088/2634-4386/ad2d5c |
| work_keys_str_mv | AT joshuarobertson vcselbasedphotonicspikingneuralnetworksforultrafastdetectionandtracking AT paulkirkland vcselbasedphotonicspikingneuralnetworksforultrafastdetectionandtracking AT gaetanodicaterina vcselbasedphotonicspikingneuralnetworksforultrafastdetectionandtracking AT antoniohurtado vcselbasedphotonicspikingneuralnetworksforultrafastdetectionandtracking |