A Novel HMM Decoding Algorithm Permitting Long-Term Dependencies and Its Application to Handwritten Word Recognition

Volkmar Frinken; Ryosuke Kakisako; Seiichi Uchida

doi:10.1109/ICFHR.2014.29

A Novel HMM Decoding Algorithm Permitting Long-Term Dependencies and Its Application to Handwritten Word Recognition

Volkmar Frinken, Ryosuke Kakisako, Seiichi Uchida

Real world robotics

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

1 Citation (Scopus)

Abstract

A new decoding for hidden Markov models is presented. As opposed to the commonly used Viterbi algorithm, it is based on the Min-Cut/Max-Flow algorithm instead of dynamic programming. Therefore non-Markovian long-term dependencies can easily be added to influence the decoding path while still finding the optimal decoding in polynomial time. We demonstrate through an experimental evaluation how these constraints can be used to improve an HMM-based handwritten word recognition system that model words via linear character-HMM by restricting the length of each character.

Original language	English
Title of host publication	Proceedings - 14th International Conference on Frontiers in Handwriting Recognition, ICFHR 2014
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	128-133
Number of pages	6
ISBN (Electronic)	9781479943340
DOIs	https://doi.org/10.1109/ICFHR.2014.29
Publication status	Published - Dec 9 2014
Event	14th International Conference on Frontiers in Handwriting Recognition, ICFHR 2014 - Hersonissos, Crete Island, Greece Duration: Sept 1 2014 → Sept 4 2014

Publication series

Name	Proceedings of International Conference on Frontiers in Handwriting Recognition, ICFHR
Volume	2014-December
ISSN (Print)	2167-6445
ISSN (Electronic)	2167-6453

Other

Other	14th International Conference on Frontiers in Handwriting Recognition, ICFHR 2014
Country/Territory	Greece
City	Hersonissos, Crete Island
Period	9/1/14 → 9/4/14

All Science Journal Classification (ASJC) codes

Computer Science Applications
Computer Vision and Pattern Recognition

Access to Document

10.1109/ICFHR.2014.29

Cite this

Frinken, V., Kakisako, R., & Uchida, S. (2014). A Novel HMM Decoding Algorithm Permitting Long-Term Dependencies and Its Application to Handwritten Word Recognition. In Proceedings - 14th International Conference on Frontiers in Handwriting Recognition, ICFHR 2014 (pp. 128-133). [6981008] (Proceedings of International Conference on Frontiers in Handwriting Recognition, ICFHR; Vol. 2014-December). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICFHR.2014.29

A Novel HMM Decoding Algorithm Permitting Long-Term Dependencies and Its Application to Handwritten Word Recognition. / Frinken, Volkmar; Kakisako, Ryosuke; Uchida, Seiichi.
Proceedings - 14th International Conference on Frontiers in Handwriting Recognition, ICFHR 2014. Institute of Electrical and Electronics Engineers Inc., 2014. p. 128-133 6981008 (Proceedings of International Conference on Frontiers in Handwriting Recognition, ICFHR; Vol. 2014-December).

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

Frinken, V, Kakisako, R & Uchida, S 2014, A Novel HMM Decoding Algorithm Permitting Long-Term Dependencies and Its Application to Handwritten Word Recognition. in Proceedings - 14th International Conference on Frontiers in Handwriting Recognition, ICFHR 2014., 6981008, Proceedings of International Conference on Frontiers in Handwriting Recognition, ICFHR, vol. 2014-December, Institute of Electrical and Electronics Engineers Inc., pp. 128-133, 14th International Conference on Frontiers in Handwriting Recognition, ICFHR 2014, Hersonissos, Crete Island, Greece, 9/1/14. https://doi.org/10.1109/ICFHR.2014.29

Frinken V, Kakisako R, Uchida S. A Novel HMM Decoding Algorithm Permitting Long-Term Dependencies and Its Application to Handwritten Word Recognition. In Proceedings - 14th International Conference on Frontiers in Handwriting Recognition, ICFHR 2014. Institute of Electrical and Electronics Engineers Inc. 2014. p. 128-133. 6981008. (Proceedings of International Conference on Frontiers in Handwriting Recognition, ICFHR). doi: 10.1109/ICFHR.2014.29

Frinken, Volkmar ; Kakisako, Ryosuke ; Uchida, Seiichi. / A Novel HMM Decoding Algorithm Permitting Long-Term Dependencies and Its Application to Handwritten Word Recognition. Proceedings - 14th International Conference on Frontiers in Handwriting Recognition, ICFHR 2014. Institute of Electrical and Electronics Engineers Inc., 2014. pp. 128-133 (Proceedings of International Conference on Frontiers in Handwriting Recognition, ICFHR).

@inproceedings{d619574288ef4109830138b7c03e28f6,

title = "A Novel HMM Decoding Algorithm Permitting Long-Term Dependencies and Its Application to Handwritten Word Recognition",

abstract = "A new decoding for hidden Markov models is presented. As opposed to the commonly used Viterbi algorithm, it is based on the Min-Cut/Max-Flow algorithm instead of dynamic programming. Therefore non-Markovian long-term dependencies can easily be added to influence the decoding path while still finding the optimal decoding in polynomial time. We demonstrate through an experimental evaluation how these constraints can be used to improve an HMM-based handwritten word recognition system that model words via linear character-HMM by restricting the length of each character.",

author = "Volkmar Frinken and Ryosuke Kakisako and Seiichi Uchida",

note = "Publisher Copyright: {\textcopyright} 2014 IEEE.; 14th International Conference on Frontiers in Handwriting Recognition, ICFHR 2014 ; Conference date: 01-09-2014 Through 04-09-2014",

year = "2014",

month = dec,

day = "9",

doi = "10.1109/ICFHR.2014.29",

language = "English",

series = "Proceedings of International Conference on Frontiers in Handwriting Recognition, ICFHR",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "128--133",

booktitle = "Proceedings - 14th International Conference on Frontiers in Handwriting Recognition, ICFHR 2014",

address = "United States",

}

TY - GEN

T1 - A Novel HMM Decoding Algorithm Permitting Long-Term Dependencies and Its Application to Handwritten Word Recognition

AU - Frinken, Volkmar

AU - Kakisako, Ryosuke

AU - Uchida, Seiichi

PY - 2014/12/9

Y1 - 2014/12/9

N2 - A new decoding for hidden Markov models is presented. As opposed to the commonly used Viterbi algorithm, it is based on the Min-Cut/Max-Flow algorithm instead of dynamic programming. Therefore non-Markovian long-term dependencies can easily be added to influence the decoding path while still finding the optimal decoding in polynomial time. We demonstrate through an experimental evaluation how these constraints can be used to improve an HMM-based handwritten word recognition system that model words via linear character-HMM by restricting the length of each character.

AB - A new decoding for hidden Markov models is presented. As opposed to the commonly used Viterbi algorithm, it is based on the Min-Cut/Max-Flow algorithm instead of dynamic programming. Therefore non-Markovian long-term dependencies can easily be added to influence the decoding path while still finding the optimal decoding in polynomial time. We demonstrate through an experimental evaluation how these constraints can be used to improve an HMM-based handwritten word recognition system that model words via linear character-HMM by restricting the length of each character.

UR - http://www.scopus.com/inward/record.url?scp=84942236657&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84942236657&partnerID=8YFLogxK

U2 - 10.1109/ICFHR.2014.29

DO - 10.1109/ICFHR.2014.29

M3 - Conference contribution

AN - SCOPUS:84942236657

T3 - Proceedings of International Conference on Frontiers in Handwriting Recognition, ICFHR

SP - 128

EP - 133

BT - Proceedings - 14th International Conference on Frontiers in Handwriting Recognition, ICFHR 2014

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 14th International Conference on Frontiers in Handwriting Recognition, ICFHR 2014

Y2 - 1 September 2014 through 4 September 2014

ER -

A Novel HMM Decoding Algorithm Permitting Long-Term Dependencies and Its Application to Handwritten Word Recognition

Abstract

Publication series

Other

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this