A biologically inspired spiking model of visual processing for image feature detection

Research output: Contribution to journalArticle

20 Citations (Scopus)

Abstract

To enable fast reliable feature matching or tracking in scenes, features need to be discrete and meaningful, and hence edge or corner features, commonly called interest points are often used for this purpose. Experimental research has illustrated that biological vision systems use neuronal circuits to extract particular features such as edges or corners from visual scenes. Inspired by this biological behaviour, this paper proposes a biologically inspired spiking neural network for the purpose of image feature extraction. Standard digital images are processed and converted to spikes in a manner similar to the processing that transforms light into spikes in the retina. Using a hierarchical spiking network, various types of biologically inspired receptive fields are used to extract progressively complex image features. The performance of the network is assessed by examining the repeatability of extracted features with visual results presented using both synthetic and real images.
LanguageEnglish
JournalNeurocomputing
VolumeX
DOIs
Publication statusPublished - 7 Feb 2015

Fingerprint

Complex Mixtures
Retina
Feature extraction
Neural networks
Light
Networks (circuits)
Processing
Research

Cite this

@article{aa68bb7c76714f81a2da58a4082e9eb2,
title = "A biologically inspired spiking model of visual processing for image feature detection",
abstract = "To enable fast reliable feature matching or tracking in scenes, features need to be discrete and meaningful, and hence edge or corner features, commonly called interest points are often used for this purpose. Experimental research has illustrated that biological vision systems use neuronal circuits to extract particular features such as edges or corners from visual scenes. Inspired by this biological behaviour, this paper proposes a biologically inspired spiking neural network for the purpose of image feature extraction. Standard digital images are processed and converted to spikes in a manner similar to the processing that transforms light into spikes in the retina. Using a hierarchical spiking network, various types of biologically inspired receptive fields are used to extract progressively complex image features. The performance of the network is assessed by examining the repeatability of extracted features with visual results presented using both synthetic and real images.",
author = "D Kerr and TM McGinnity and SA Coleman and M Clogenson",
year = "2015",
month = "2",
day = "7",
doi = "10.1016/j.neucom.2015.01.011",
language = "English",
volume = "X",
journal = "Neurocomputing",
issn = "0925-2312",
publisher = "Elsevier",

}

A biologically inspired spiking model of visual processing for image feature detection. / Kerr, D; McGinnity, TM; Coleman, SA; Clogenson, M.

In: Neurocomputing, Vol. X, 07.02.2015.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A biologically inspired spiking model of visual processing for image feature detection

AU - Kerr, D

AU - McGinnity, TM

AU - Coleman, SA

AU - Clogenson, M

PY - 2015/2/7

Y1 - 2015/2/7

N2 - To enable fast reliable feature matching or tracking in scenes, features need to be discrete and meaningful, and hence edge or corner features, commonly called interest points are often used for this purpose. Experimental research has illustrated that biological vision systems use neuronal circuits to extract particular features such as edges or corners from visual scenes. Inspired by this biological behaviour, this paper proposes a biologically inspired spiking neural network for the purpose of image feature extraction. Standard digital images are processed and converted to spikes in a manner similar to the processing that transforms light into spikes in the retina. Using a hierarchical spiking network, various types of biologically inspired receptive fields are used to extract progressively complex image features. The performance of the network is assessed by examining the repeatability of extracted features with visual results presented using both synthetic and real images.

AB - To enable fast reliable feature matching or tracking in scenes, features need to be discrete and meaningful, and hence edge or corner features, commonly called interest points are often used for this purpose. Experimental research has illustrated that biological vision systems use neuronal circuits to extract particular features such as edges or corners from visual scenes. Inspired by this biological behaviour, this paper proposes a biologically inspired spiking neural network for the purpose of image feature extraction. Standard digital images are processed and converted to spikes in a manner similar to the processing that transforms light into spikes in the retina. Using a hierarchical spiking network, various types of biologically inspired receptive fields are used to extract progressively complex image features. The performance of the network is assessed by examining the repeatability of extracted features with visual results presented using both synthetic and real images.

U2 - 10.1016/j.neucom.2015.01.011

DO - 10.1016/j.neucom.2015.01.011

M3 - Article

VL - X

JO - Neurocomputing

T2 - Neurocomputing

JF - Neurocomputing

SN - 0925-2312

ER -