Analysis of neural networks for edge detection

B J vanderZwaag; C H Slump; Lambert Spaanenburg

Analysis of neural networks for edge detection

B J vanderZwaag, C H Slump, Lambert Spaanenburg

External Organization - Unknown

Research output: Chapter in Book/Report/Conference proceeding › Paper in conference proceeding › peer-review

Abstract

This paper illustrates a novel method to analyze artificial neural networks so as to gain insight into
their internal functionality. To this purpose, the elements of a feedforward-backpropagation neural network, that has been trained to detect edges in images, are described in terms of differential operators of various orders and with various angles of operation.

Original language	English
Title of host publication	Proceedings of the ProRISC Workshop on Circuits, Systems and Signal Processing
Publisher	STW Technology Foundation
Pages	580-586
ISBN (Print)	90-73461-33-2
Publication status	Published - 2002
Externally published	Yes
Event	ProRisc - Veldhoven, The Netherlands, Netherlands Duration: 2002 Nov 28 → 2002 Nov 29

Conference

Conference	ProRisc
Country/Territory	Netherlands
Period	2002/11/28 → 2002/11/29

Subject classification (UKÄ)

Electrical Engineering, Electronic Engineering, Information Engineering

Free keywords

rule extraction
Neural networks
digital image processing
edge detection
gradient filters.

Cite this

@inproceedings{2760a322eb3e4cc5a49740981cd7753f,

title = "Analysis of neural networks for edge detection",

abstract = "This paper illustrates a novel method to analyze artificial neural networks so as to gain insight into their internal functionality. To this purpose, the elements of a feedforward-backpropagation neural network, that has been trained to detect edges in images, are described in terms of differential operators of various orders and with various angles of operation.",

keywords = "rule extraction, Neural networks, digital image processing, edge detection, gradient filters.",

author = "vanderZwaag, {B J} and Slump, {C H} and Lambert Spaanenburg",

year = "2002",

language = "English",

isbn = "90-73461-33-2",

pages = "580--586",

booktitle = "Proceedings of the ProRISC Workshop on Circuits, Systems and Signal Processing",

publisher = "STW Technology Foundation",

note = "ProRisc ; Conference date: 28-11-2002 Through 29-11-2002",

}

TY - GEN

T1 - Analysis of neural networks for edge detection

AU - vanderZwaag, B J

AU - Slump, C H

AU - Spaanenburg, Lambert

PY - 2002

Y1 - 2002

N2 - This paper illustrates a novel method to analyze artificial neural networks so as to gain insight into their internal functionality. To this purpose, the elements of a feedforward-backpropagation neural network, that has been trained to detect edges in images, are described in terms of differential operators of various orders and with various angles of operation.

AB - This paper illustrates a novel method to analyze artificial neural networks so as to gain insight into their internal functionality. To this purpose, the elements of a feedforward-backpropagation neural network, that has been trained to detect edges in images, are described in terms of differential operators of various orders and with various angles of operation.

KW - rule extraction

KW - Neural networks

KW - digital image processing

KW - edge detection

KW - gradient filters.

M3 - Paper in conference proceeding

SN - 90-73461-33-2

SP - 580

EP - 586

BT - Proceedings of the ProRISC Workshop on Circuits, Systems and Signal Processing

PB - STW Technology Foundation

T2 - ProRisc

Y2 - 28 November 2002 through 29 November 2002

ER -

Analysis of neural networks for edge detection

Abstract

Conference

Subject classification (UKÄ)

Free keywords

Fingerprint

Cite this