Hybrid pattern recognition system capable of self-modification

Charles W. Glover; Nageswara S.V. Rao; E. M. Oblow

doi:10.1007/3-540-54563-8_97

Hybrid pattern recognition system capable of self-modification

Charles W. Glover
, Nageswara S.V. Rao
, E. M. Oblow

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

Abstract

Systems capable of recognizing and learning two-dimensional patterns can be used in imaging systems and robotic perception systems. The symbolic and neuromorphic methods for pattern processing problems of this type are complementary in character. We present a hybrid system that utilizes components of symbolic and neuromorphic type; we employ two hybrid components that simultaneously operate up on the same data to produce hypotheses about the data. To resolve the potential conflicts in these hypotheses, we propose a method that learns a combination rule based on a set of examples. We employ the method of empirical risk minimization that does not require knowledge about the error probability distributions of the modules. We are building a prototype system to recognize control panels using a vision system.

Original language	English
Title of host publication	Methodologies for Intelligent Systems - 6th International Symposium, ISMIS 1991, Proceedings
Editors	Zbigniew W. Ras, Maria Zemankova
Publisher	Springer Verlag
Pages	338-347
Number of pages	10
ISBN (Print)	9783540545637
DOIs	https://doi.org/10.1007/3-540-54563-8_97
State	Published - 1991
Externally published	Yes
Event	6th International Symposium on Methodologies for Intelligent Systems, ISMIS 1991 - Charlotte , United States Duration: Oct 16 1991 → Oct 19 1991

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	542 LNAI Part F2
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	6th International Symposium on Methodologies for Intelligent Systems, ISMIS 1991
Country/Territory	United States
City	Charlotte
Period	10/16/91 → 10/19/91

Funding

The main difficulties in building such a system arise due to (a) sensor noise, and (b) structural changes in a control panel that are legitimately allowed, e.g., sliding a handle of t Partially funded by Virginia's Center for Innovative Technology under grant #INF-90-015, the Deparlment of Energy through Oak Ridge National Laboratoryo perated by Martin Marietta Energy Systems, Inc., under the contract #19X-SE043V, and by Old Dominion University Summer Faculty Fellowship for 1991.

Access to Document

10.1007/3-540-54563-8_97

Cite this

Glover, C. W., Rao, N. S. V., & Oblow, E. M. (1991). Hybrid pattern recognition system capable of self-modification. In Z. W. Ras, & M. Zemankova (Eds.), Methodologies for Intelligent Systems - 6th International Symposium, ISMIS 1991, Proceedings (pp. 338-347). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 542 LNAI Part F2). Springer Verlag. https://doi.org/10.1007/3-540-54563-8_97

Glover, Charles W. ; Rao, Nageswara S.V. ; Oblow, E. M. / Hybrid pattern recognition system capable of self-modification. Methodologies for Intelligent Systems - 6th International Symposium, ISMIS 1991, Proceedings. editor / Zbigniew W. Ras ; Maria Zemankova. Springer Verlag, 1991. pp. 338-347 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{1aef2f6114f14f1089cfa672f1bdbfdf,

title = "Hybrid pattern recognition system capable of self-modification",

abstract = "Systems capable of recognizing and learning two-dimensional patterns can be used in imaging systems and robotic perception systems. The symbolic and neuromorphic methods for pattern processing problems of this type are complementary in character. We present a hybrid system that utilizes components of symbolic and neuromorphic type; we employ two hybrid components that simultaneously operate up on the same data to produce hypotheses about the data. To resolve the potential conflicts in these hypotheses, we propose a method that learns a combination rule based on a set of examples. We employ the method of empirical risk minimization that does not require knowledge about the error probability distributions of the modules. We are building a prototype system to recognize control panels using a vision system.",

author = "Glover, \{Charles W.\} and Rao, \{Nageswara S.V.\} and Oblow, \{E. M.\}",

note = "Publisher Copyright: {\textcopyright} Springer-Verlag Berlin Heidelberg 1991.; 6th International Symposium on Methodologies for Intelligent Systems, ISMIS 1991 ; Conference date: 16-10-1991 Through 19-10-1991",

year = "1991",

doi = "10.1007/3-540-54563-8\_97",

language = "English",

isbn = "9783540545637",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "338--347",

editor = "Ras, \{Zbigniew W.\} and Maria Zemankova",

booktitle = "Methodologies for Intelligent Systems - 6th International Symposium, ISMIS 1991, Proceedings",

}

Glover, CW, Rao, NSV & Oblow, EM 1991, Hybrid pattern recognition system capable of self-modification. in ZW Ras & M Zemankova (eds), Methodologies for Intelligent Systems - 6th International Symposium, ISMIS 1991, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 542 LNAI Part F2, Springer Verlag, pp. 338-347, 6th International Symposium on Methodologies for Intelligent Systems, ISMIS 1991, Charlotte , United States, 10/16/91. https://doi.org/10.1007/3-540-54563-8_97

Hybrid pattern recognition system capable of self-modification. / Glover, Charles W.; Rao, Nageswara S.V.; Oblow, E. M.
Methodologies for Intelligent Systems - 6th International Symposium, ISMIS 1991, Proceedings. ed. / Zbigniew W. Ras; Maria Zemankova. Springer Verlag, 1991. p. 338-347 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 542 LNAI Part F2).

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

TY - GEN

T1 - Hybrid pattern recognition system capable of self-modification

AU - Glover, Charles W.

AU - Rao, Nageswara S.V.

AU - Oblow, E. M.

N1 - Publisher Copyright: © Springer-Verlag Berlin Heidelberg 1991.

PY - 1991

Y1 - 1991

N2 - Systems capable of recognizing and learning two-dimensional patterns can be used in imaging systems and robotic perception systems. The symbolic and neuromorphic methods for pattern processing problems of this type are complementary in character. We present a hybrid system that utilizes components of symbolic and neuromorphic type; we employ two hybrid components that simultaneously operate up on the same data to produce hypotheses about the data. To resolve the potential conflicts in these hypotheses, we propose a method that learns a combination rule based on a set of examples. We employ the method of empirical risk minimization that does not require knowledge about the error probability distributions of the modules. We are building a prototype system to recognize control panels using a vision system.

AB - Systems capable of recognizing and learning two-dimensional patterns can be used in imaging systems and robotic perception systems. The symbolic and neuromorphic methods for pattern processing problems of this type are complementary in character. We present a hybrid system that utilizes components of symbolic and neuromorphic type; we employ two hybrid components that simultaneously operate up on the same data to produce hypotheses about the data. To resolve the potential conflicts in these hypotheses, we propose a method that learns a combination rule based on a set of examples. We employ the method of empirical risk minimization that does not require knowledge about the error probability distributions of the modules. We are building a prototype system to recognize control panels using a vision system.

UR - https://www.scopus.com/pages/publications/85037072477

U2 - 10.1007/3-540-54563-8_97

DO - 10.1007/3-540-54563-8_97

M3 - Conference contribution

AN - SCOPUS:85037072477

SN - 9783540545637

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 338

EP - 347

BT - Methodologies for Intelligent Systems - 6th International Symposium, ISMIS 1991, Proceedings

A2 - Ras, Zbigniew W.

A2 - Zemankova, Maria

PB - Springer Verlag

T2 - 6th International Symposium on Methodologies for Intelligent Systems, ISMIS 1991

Y2 - 16 October 1991 through 19 October 1991

ER -

Glover CW, Rao NSV, Oblow EM. Hybrid pattern recognition system capable of self-modification. In Ras ZW, Zemankova M, editors, Methodologies for Intelligent Systems - 6th International Symposium, ISMIS 1991, Proceedings. Springer Verlag. 1991. p. 338-347. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/3-540-54563-8_97

Hybrid pattern recognition system capable of self-modification

Abstract

Publication series

Conference

Funding

Access to Document

Other files and links

Fingerprint

Cite this