Kernel based principal component for recognizing handwritten numbers

Xiaoxiao Song; Songhua Xu; Willard L. Miranker

doi:10.1109/IJCNN.2010.5596733

Kernel based principal component for recognizing handwritten numbers

Xiaoxiao Song, Songhua Xu, Willard L. Miranker

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

Abstract

We implement kernel-based principal component analysis to recognize handwritten numbers. Then we analyze the relationship of each numeral type's eigenvalue and eigenvector with its recognition rate. We also present a modified algorithm to improve the robustness as well as the efficiency of the recognition method by employing the secondary training and detection methods from the perspective of nature of kernel function. This method can solve the problem of low recognition rate of a small number of scribbled characters at both low time cost and space complexity. Experiments using 1000 to 5000 test samples all show that our method can achieve 97.8% to 99.0% recognition accuracy.

Original language	English (US)
Title of host publication	2010 IEEE World Congress on Computational Intelligence, WCCI 2010 - 2010 International Joint Conference on Neural Networks, IJCNN 2010
DOIs	https://doi.org/10.1109/IJCNN.2010.5596733
State	Published - 2010
Externally published	Yes
Event	2010 6th IEEE World Congress on Computational Intelligence, WCCI 2010 - 2010 International Joint Conference on Neural Networks, IJCNN 2010 - Barcelona, Spain Duration: Jul 18 2010 → Jul 23 2010

Publication series

Name	Proceedings of the International Joint Conference on Neural Networks

Other

Other	2010 6th IEEE World Congress on Computational Intelligence, WCCI 2010 - 2010 International Joint Conference on Neural Networks, IJCNN 2010
Country/Territory	Spain
City	Barcelona
Period	7/18/10 → 7/23/10

All Science Journal Classification (ASJC) codes

Software
Artificial Intelligence

Access to Document

10.1109/IJCNN.2010.5596733

Cite this

Song, X., Xu, S., & Miranker, W. L. (2010). Kernel based principal component for recognizing handwritten numbers. In 2010 IEEE World Congress on Computational Intelligence, WCCI 2010 - 2010 International Joint Conference on Neural Networks, IJCNN 2010 [5596733] (Proceedings of the International Joint Conference on Neural Networks). https://doi.org/10.1109/IJCNN.2010.5596733

@inproceedings{1d49a45fdc7c48d699d25813df9293a9,

title = "Kernel based principal component for recognizing handwritten numbers",

abstract = "We implement kernel-based principal component analysis to recognize handwritten numbers. Then we analyze the relationship of each numeral type's eigenvalue and eigenvector with its recognition rate. We also present a modified algorithm to improve the robustness as well as the efficiency of the recognition method by employing the secondary training and detection methods from the perspective of nature of kernel function. This method can solve the problem of low recognition rate of a small number of scribbled characters at both low time cost and space complexity. Experiments using 1000 to 5000 test samples all show that our method can achieve 97.8% to 99.0% recognition accuracy.",

author = "Xiaoxiao Song and Songhua Xu and Miranker, {Willard L.}",

year = "2010",

doi = "10.1109/IJCNN.2010.5596733",

language = "English (US)",

isbn = "9781424469178",

series = "Proceedings of the International Joint Conference on Neural Networks",

booktitle = "2010 IEEE World Congress on Computational Intelligence, WCCI 2010 - 2010 International Joint Conference on Neural Networks, IJCNN 2010",

note = "2010 6th IEEE World Congress on Computational Intelligence, WCCI 2010 - 2010 International Joint Conference on Neural Networks, IJCNN 2010 ; Conference date: 18-07-2010 Through 23-07-2010",

}

Song, X, Xu, S & Miranker, WL 2010, Kernel based principal component for recognizing handwritten numbers. in 2010 IEEE World Congress on Computational Intelligence, WCCI 2010 - 2010 International Joint Conference on Neural Networks, IJCNN 2010., 5596733, Proceedings of the International Joint Conference on Neural Networks, 2010 6th IEEE World Congress on Computational Intelligence, WCCI 2010 - 2010 International Joint Conference on Neural Networks, IJCNN 2010, Barcelona, Spain, 7/18/10. https://doi.org/10.1109/IJCNN.2010.5596733

Kernel based principal component for recognizing handwritten numbers. / Song, Xiaoxiao; Xu, Songhua; Miranker, Willard L.
2010 IEEE World Congress on Computational Intelligence, WCCI 2010 - 2010 International Joint Conference on Neural Networks, IJCNN 2010. 2010. 5596733 (Proceedings of the International Joint Conference on Neural Networks).

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

TY - GEN

T1 - Kernel based principal component for recognizing handwritten numbers

AU - Song, Xiaoxiao

AU - Xu, Songhua

AU - Miranker, Willard L.

PY - 2010

Y1 - 2010

N2 - We implement kernel-based principal component analysis to recognize handwritten numbers. Then we analyze the relationship of each numeral type's eigenvalue and eigenvector with its recognition rate. We also present a modified algorithm to improve the robustness as well as the efficiency of the recognition method by employing the secondary training and detection methods from the perspective of nature of kernel function. This method can solve the problem of low recognition rate of a small number of scribbled characters at both low time cost and space complexity. Experiments using 1000 to 5000 test samples all show that our method can achieve 97.8% to 99.0% recognition accuracy.

AB - We implement kernel-based principal component analysis to recognize handwritten numbers. Then we analyze the relationship of each numeral type's eigenvalue and eigenvector with its recognition rate. We also present a modified algorithm to improve the robustness as well as the efficiency of the recognition method by employing the secondary training and detection methods from the perspective of nature of kernel function. This method can solve the problem of low recognition rate of a small number of scribbled characters at both low time cost and space complexity. Experiments using 1000 to 5000 test samples all show that our method can achieve 97.8% to 99.0% recognition accuracy.

UR - http://www.scopus.com/inward/record.url?scp=79959420513&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79959420513&partnerID=8YFLogxK

U2 - 10.1109/IJCNN.2010.5596733

DO - 10.1109/IJCNN.2010.5596733

M3 - Conference contribution

AN - SCOPUS:79959420513

SN - 9781424469178

T3 - Proceedings of the International Joint Conference on Neural Networks

BT - 2010 IEEE World Congress on Computational Intelligence, WCCI 2010 - 2010 International Joint Conference on Neural Networks, IJCNN 2010

T2 - 2010 6th IEEE World Congress on Computational Intelligence, WCCI 2010 - 2010 International Joint Conference on Neural Networks, IJCNN 2010

Y2 - 18 July 2010 through 23 July 2010

ER -

Kernel based principal component for recognizing handwritten numbers

Abstract

Publication series

Other

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this