An edge detection algorithm using multi-state adalines

Joon K. Paik; James C. Brailean; Aggelos K. Katsaggelos

doi:10.1016/0031-3203(92)90122-Y

An edge detection algorithm using multi-state adalines

Joon K. Paik, James C. Brailean, Aggelos K. Katsaggelos^*

^*Corresponding author for this work

Electrical and Computer Engineering

Research output: Contribution to journal › Article › peer-review

7 Scopus citations

Abstract

An edge detection algorithm using multi-state adaptive linear neurons (ADALINES) is presented. Although the tri-state ADALINE is only considered in this work, general multi-state input vectors with extreme values are shown to be linearly separable from the rest of the vectors with the same dimension. The input state of each ADALINE is defined using the local mean in a predefined mask. In addition to the binary input states ± 1, the 0 input state is introduced for controlling the noise effect. If the input pattern matches one of the predefined edge patterns, the corresponding pixel is detected as an edge pixel. Experimental results are shown where the proposed detector is compared with both the Canny and LOG edge detectors.

Original language	English (US)
Pages (from-to)	1495-1504
Number of pages	10
Journal	Pattern Recognition
Volume	25
Issue number	12
DOIs	https://doi.org/10.1016/0031-3203(92)90122-Y
State	Published - Dec 1992

Keywords

Edge detection
Linear neural networks
Pattern recognition

ASJC Scopus subject areas

Software
Signal Processing
Computer Vision and Pattern Recognition
Artificial Intelligence

Access to Document

10.1016/0031-3203(92)90122-Y

Fingerprint Dive into the research topics of 'An edge detection algorithm using multi-state adalines'. Together they form a unique fingerprint.

Cite this

@article{9bf0bde1f6064241947ab4e12d6f36e3,

title = "An edge detection algorithm using multi-state adalines",

abstract = "An edge detection algorithm using multi-state adaptive linear neurons (ADALINES) is presented. Although the tri-state ADALINE is only considered in this work, general multi-state input vectors with extreme values are shown to be linearly separable from the rest of the vectors with the same dimension. The input state of each ADALINE is defined using the local mean in a predefined mask. In addition to the binary input states ± 1, the 0 input state is introduced for controlling the noise effect. If the input pattern matches one of the predefined edge patterns, the corresponding pixel is detected as an edge pixel. Experimental results are shown where the proposed detector is compared with both the Canny and LOG edge detectors.",

keywords = "Edge detection, Linear neural networks, Pattern recognition",

author = "Paik, {Joon K.} and Brailean, {James C.} and Katsaggelos, {Aggelos K.}",

year = "1992",

month = dec,

doi = "10.1016/0031-3203(92)90122-Y",

language = "English (US)",

volume = "25",

pages = "1495--1504",

journal = "Pattern Recognition",

issn = "0031-3203",

publisher = "Elsevier Limited",

number = "12",

}

TY - JOUR

T1 - An edge detection algorithm using multi-state adalines

AU - Paik, Joon K.

AU - Brailean, James C.

AU - Katsaggelos, Aggelos K.

PY - 1992/12

Y1 - 1992/12

N2 - An edge detection algorithm using multi-state adaptive linear neurons (ADALINES) is presented. Although the tri-state ADALINE is only considered in this work, general multi-state input vectors with extreme values are shown to be linearly separable from the rest of the vectors with the same dimension. The input state of each ADALINE is defined using the local mean in a predefined mask. In addition to the binary input states ± 1, the 0 input state is introduced for controlling the noise effect. If the input pattern matches one of the predefined edge patterns, the corresponding pixel is detected as an edge pixel. Experimental results are shown where the proposed detector is compared with both the Canny and LOG edge detectors.

AB - An edge detection algorithm using multi-state adaptive linear neurons (ADALINES) is presented. Although the tri-state ADALINE is only considered in this work, general multi-state input vectors with extreme values are shown to be linearly separable from the rest of the vectors with the same dimension. The input state of each ADALINE is defined using the local mean in a predefined mask. In addition to the binary input states ± 1, the 0 input state is introduced for controlling the noise effect. If the input pattern matches one of the predefined edge patterns, the corresponding pixel is detected as an edge pixel. Experimental results are shown where the proposed detector is compared with both the Canny and LOG edge detectors.

KW - Edge detection

KW - Linear neural networks

KW - Pattern recognition

UR - http://www.scopus.com/inward/record.url?scp=0026959140&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0026959140&partnerID=8YFLogxK

U2 - 10.1016/0031-3203(92)90122-Y

DO - 10.1016/0031-3203(92)90122-Y

M3 - Article

AN - SCOPUS:0026959140

VL - 25

SP - 1495

EP - 1504

JO - Pattern Recognition

JF - Pattern Recognition

SN - 0031-3203

IS - 12

ER -