Cargando…

A Motion-Based Feature for Event-Based Pattern Recognition

This paper introduces an event-based luminance-free feature from the output of asynchronous event-based neuromorphic retinas. The feature consists in mapping the distribution of the optical flow along the contours of the moving objects in the visual scene into a matrix. Asynchronous event-based neur...

Descripción completa

Detalles Bibliográficos
Autores principales: Clady, Xavier, Maro, Jean-Matthieu, Barré, Sébastien, Benosman, Ryad B.
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Frontiers Media S.A. 2017
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5209354/
https://www.ncbi.nlm.nih.gov/pubmed/28101001
http://dx.doi.org/10.3389/fnins.2016.00594
_version_ 1782490724266147840
author Clady, Xavier
Maro, Jean-Matthieu
Barré, Sébastien
Benosman, Ryad B.
author_facet Clady, Xavier
Maro, Jean-Matthieu
Barré, Sébastien
Benosman, Ryad B.
author_sort Clady, Xavier
collection PubMed
description This paper introduces an event-based luminance-free feature from the output of asynchronous event-based neuromorphic retinas. The feature consists in mapping the distribution of the optical flow along the contours of the moving objects in the visual scene into a matrix. Asynchronous event-based neuromorphic retinas are composed of autonomous pixels, each of them asynchronously generating “spiking” events that encode relative changes in pixels' illumination at high temporal resolutions. The optical flow is computed at each event, and is integrated locally or globally in a speed and direction coordinate frame based grid, using speed-tuned temporal kernels. The latter ensures that the resulting feature equitably represents the distribution of the normal motion along the current moving edges, whatever their respective dynamics. The usefulness and the generality of the proposed feature are demonstrated in pattern recognition applications: local corner detection and global gesture recognition.
format Online
Article
Text
id pubmed-5209354
institution National Center for Biotechnology Information
language English
publishDate 2017
publisher Frontiers Media S.A.
record_format MEDLINE/PubMed
spelling pubmed-52093542017-01-18 A Motion-Based Feature for Event-Based Pattern Recognition Clady, Xavier Maro, Jean-Matthieu Barré, Sébastien Benosman, Ryad B. Front Neurosci Neuroscience This paper introduces an event-based luminance-free feature from the output of asynchronous event-based neuromorphic retinas. The feature consists in mapping the distribution of the optical flow along the contours of the moving objects in the visual scene into a matrix. Asynchronous event-based neuromorphic retinas are composed of autonomous pixels, each of them asynchronously generating “spiking” events that encode relative changes in pixels' illumination at high temporal resolutions. The optical flow is computed at each event, and is integrated locally or globally in a speed and direction coordinate frame based grid, using speed-tuned temporal kernels. The latter ensures that the resulting feature equitably represents the distribution of the normal motion along the current moving edges, whatever their respective dynamics. The usefulness and the generality of the proposed feature are demonstrated in pattern recognition applications: local corner detection and global gesture recognition. Frontiers Media S.A. 2017-01-04 /pmc/articles/PMC5209354/ /pubmed/28101001 http://dx.doi.org/10.3389/fnins.2016.00594 Text en Copyright © 2017 Clady, Maro, Barré and Benosman. http://creativecommons.org/licenses/by/4.0/ This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
spellingShingle Neuroscience
Clady, Xavier
Maro, Jean-Matthieu
Barré, Sébastien
Benosman, Ryad B.
A Motion-Based Feature for Event-Based Pattern Recognition
title A Motion-Based Feature for Event-Based Pattern Recognition
title_full A Motion-Based Feature for Event-Based Pattern Recognition
title_fullStr A Motion-Based Feature for Event-Based Pattern Recognition
title_full_unstemmed A Motion-Based Feature for Event-Based Pattern Recognition
title_short A Motion-Based Feature for Event-Based Pattern Recognition
title_sort motion-based feature for event-based pattern recognition
topic Neuroscience
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5209354/
https://www.ncbi.nlm.nih.gov/pubmed/28101001
http://dx.doi.org/10.3389/fnins.2016.00594
work_keys_str_mv AT cladyxavier amotionbasedfeatureforeventbasedpatternrecognition
AT marojeanmatthieu amotionbasedfeatureforeventbasedpatternrecognition
AT barresebastien amotionbasedfeatureforeventbasedpatternrecognition
AT benosmanryadb amotionbasedfeatureforeventbasedpatternrecognition
AT cladyxavier motionbasedfeatureforeventbasedpatternrecognition
AT marojeanmatthieu motionbasedfeatureforeventbasedpatternrecognition
AT barresebastien motionbasedfeatureforeventbasedpatternrecognition
AT benosmanryadb motionbasedfeatureforeventbasedpatternrecognition