Transformation of optimized prototypes for handwritten digit recognition

Hong Yan

doi:10.1109/ICASSP.1994.389578

Transformation of optimized prototypes for handwritten digit recognition

Hong Yan^*

^*Corresponding author for this work

Research output: Journal Publications and Reviews › RGC 21 - Publication in refereed journal › peer-review

Abstract

We propose a method in this paper for handwritten digit recognition using optimized prototypes generated through learning and transformation. In this method a set of prototypes are obtained from tra.ining samples and mapped to a multi-layer neural network for optimization to improve their classification power. The new prototypes are then transformed geometrically to produce a larger set of prototypes for recognition of testing samples. The method has been verified to work well in experimental studies.

Original language	English
Article number	389578
Pages (from-to)	II625-II628
Journal	ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume	2
DOIs	https://doi.org/10.1109/ICASSP.1994.389578
Publication status	Published - 1994
Externally published	Yes
Event	Proceedings of the 1994 IEEE International Conference on Acoustics, Speech and Signal Processing. Part 2 (of 6) - Adelaide, Aust Duration: 19 Apr 1994 → 22 Apr 1994

Bibliographical note

Publication details (e.g. title, author(s), publication statuses and dates) are captured on an “AS IS” and “AS AVAILABLE” basis at the time of record harvesting from the data source. Suggestions for further amendments or supplementary information can be sent to [email protected].

Access Output

10.1109/ICASSP.1994.389578

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

@article{ff3743cd041445448c95aeea7df71146,

title = "Transformation of optimized prototypes for handwritten digit recognition",

abstract = "We propose a method in this paper for handwritten digit recognition using optimized prototypes generated through learning and transformation. In this method a set of prototypes are obtained from tra.ining samples and mapped to a multi-layer neural network for optimization to improve their classification power. The new prototypes are then transformed geometrically to produce a larger set of prototypes for recognition of testing samples. The method has been verified to work well in experimental studies.",

author = "Hong Yan",

note = "Publication details (e.g. title, author(s), publication statuses and dates) are captured on an “AS IS” and “AS AVAILABLE” basis at the time of record harvesting from the data source. Suggestions for further amendments or supplementary information can be sent to [email protected].; Proceedings of the 1994 IEEE International Conference on Acoustics, Speech and Signal Processing. Part 2 (of 6) ; Conference date: 19-04-1994 Through 22-04-1994",

year = "1994",

doi = "10.1109/ICASSP.1994.389578",

language = "English",

volume = "2",

pages = "II625--II628",

}

TY - JOUR

T1 - Transformation of optimized prototypes for handwritten digit recognition

AU - Yan, Hong

N1 - Publication details (e.g. title, author(s), publication statuses and dates) are captured on an “AS IS” and “AS AVAILABLE” basis at the time of record harvesting from the data source. Suggestions for further amendments or supplementary information can be sent to [email protected].

PY - 1994

Y1 - 1994

N2 - We propose a method in this paper for handwritten digit recognition using optimized prototypes generated through learning and transformation. In this method a set of prototypes are obtained from tra.ining samples and mapped to a multi-layer neural network for optimization to improve their classification power. The new prototypes are then transformed geometrically to produce a larger set of prototypes for recognition of testing samples. The method has been verified to work well in experimental studies.

AB - We propose a method in this paper for handwritten digit recognition using optimized prototypes generated through learning and transformation. In this method a set of prototypes are obtained from tra.ining samples and mapped to a multi-layer neural network for optimization to improve their classification power. The new prototypes are then transformed geometrically to produce a larger set of prototypes for recognition of testing samples. The method has been verified to work well in experimental studies.

UR - http://www.scopus.com/inward/record.url?scp=85079247048&partnerID=8YFLogxK

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-85079247048&origin=recordpage

U2 - 10.1109/ICASSP.1994.389578

DO - 10.1109/ICASSP.1994.389578

M3 - RGC 21 - Publication in refereed journal

SN - 1520-6149

VL - 2

SP - II625-II628

JO - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

JF - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

M1 - 389578

T2 - Proceedings of the 1994 IEEE International Conference on Acoustics, Speech and Signal Processing. Part 2 (of 6)

Y2 - 19 April 1994 through 22 April 1994

ER -

Transformation of optimized prototypes for handwritten digit recognition

Abstract

Bibliographical note

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Cite this