Handwritten and machine-printed text discrimination using a template matching approach

Mehryar Emambakhsh, Yulan He, Ian Nabney

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We propose a novel template matching approach for the discrimination of handwritten and machine-printed text. We first pre-process the scanned document images by performing denoising, circles/lines exclusion and word-block level segmentation. We then align and match characters in a flexible sized gallery with the segmented regions, using parallelised normalised cross-correlation. The experimental results over the Pattern Recognition & Image Analysis Research Lab-Natural History Museum (PRImA-NHM) dataset show remarkably high robustness of the algorithm in classifying cluttered, occluded and noisy samples, in addition to those with significant high missing data. The algorithm, which gives 84.0% classification rate with false positive rate 0.16 over the dataset, does not require training samples and generates compelling results as opposed to the training-based approaches, which have used the same benchmark.

Original languageEnglish
Title of host publicationProceedings : 12th IAPR International Workshop on Document Analysis Systems, DAS 2016
PublisherIEEE
Pages399-404
Number of pages6
ISBN (Print)978-1-5090-1792-8
DOIs
Publication statusPublished - 13 Jun 2016
Event12th IAPR International Workshop on Document Analysis Systems - Santorini, Greece
Duration: 11 Apr 201614 Apr 2016

Workshop

Workshop12th IAPR International Workshop on Document Analysis Systems
Abbreviated titleDAS 2016
CountryGreece
CitySantorini
Period11/04/1614/04/16

Fingerprint

Template matching
discrimination
pattern recognition
Museums
Image analysis
Pattern recognition
museum
exclusion
history
segmentation

Bibliographical note

-© 2016 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

Keywords

  • classification
  • handwritten
  • machine-printed
  • OCR
  • shape analysis
  • template matching

Cite this

Emambakhsh, M., He, Y., & Nabney, I. (2016). Handwritten and machine-printed text discrimination using a template matching approach. In Proceedings : 12th IAPR International Workshop on Document Analysis Systems, DAS 2016 (pp. 399-404). IEEE. https://doi.org/10.1109/DAS.2016.22
Emambakhsh, Mehryar ; He, Yulan ; Nabney, Ian. / Handwritten and machine-printed text discrimination using a template matching approach. Proceedings : 12th IAPR International Workshop on Document Analysis Systems, DAS 2016. IEEE, 2016. pp. 399-404
@inproceedings{f197b789a2514a50b5308d0d2c0983da,
title = "Handwritten and machine-printed text discrimination using a template matching approach",
abstract = "We propose a novel template matching approach for the discrimination of handwritten and machine-printed text. We first pre-process the scanned document images by performing denoising, circles/lines exclusion and word-block level segmentation. We then align and match characters in a flexible sized gallery with the segmented regions, using parallelised normalised cross-correlation. The experimental results over the Pattern Recognition & Image Analysis Research Lab-Natural History Museum (PRImA-NHM) dataset show remarkably high robustness of the algorithm in classifying cluttered, occluded and noisy samples, in addition to those with significant high missing data. The algorithm, which gives 84.0{\%} classification rate with false positive rate 0.16 over the dataset, does not require training samples and generates compelling results as opposed to the training-based approaches, which have used the same benchmark.",
keywords = "classification, handwritten, machine-printed, OCR, shape analysis, template matching",
author = "Mehryar Emambakhsh and Yulan He and Ian Nabney",
note = "-{\circledC} 2016 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.",
year = "2016",
month = "6",
day = "13",
doi = "10.1109/DAS.2016.22",
language = "English",
isbn = "978-1-5090-1792-8",
pages = "399--404",
booktitle = "Proceedings : 12th IAPR International Workshop on Document Analysis Systems, DAS 2016",
publisher = "IEEE",
address = "United States",

}

Emambakhsh, M, He, Y & Nabney, I 2016, Handwritten and machine-printed text discrimination using a template matching approach. in Proceedings : 12th IAPR International Workshop on Document Analysis Systems, DAS 2016. IEEE, pp. 399-404, 12th IAPR International Workshop on Document Analysis Systems, Santorini, Greece, 11/04/16. https://doi.org/10.1109/DAS.2016.22

Handwritten and machine-printed text discrimination using a template matching approach. / Emambakhsh, Mehryar; He, Yulan; Nabney, Ian.

Proceedings : 12th IAPR International Workshop on Document Analysis Systems, DAS 2016. IEEE, 2016. p. 399-404.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Handwritten and machine-printed text discrimination using a template matching approach

AU - Emambakhsh, Mehryar

AU - He, Yulan

AU - Nabney, Ian

N1 - -© 2016 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

PY - 2016/6/13

Y1 - 2016/6/13

N2 - We propose a novel template matching approach for the discrimination of handwritten and machine-printed text. We first pre-process the scanned document images by performing denoising, circles/lines exclusion and word-block level segmentation. We then align and match characters in a flexible sized gallery with the segmented regions, using parallelised normalised cross-correlation. The experimental results over the Pattern Recognition & Image Analysis Research Lab-Natural History Museum (PRImA-NHM) dataset show remarkably high robustness of the algorithm in classifying cluttered, occluded and noisy samples, in addition to those with significant high missing data. The algorithm, which gives 84.0% classification rate with false positive rate 0.16 over the dataset, does not require training samples and generates compelling results as opposed to the training-based approaches, which have used the same benchmark.

AB - We propose a novel template matching approach for the discrimination of handwritten and machine-printed text. We first pre-process the scanned document images by performing denoising, circles/lines exclusion and word-block level segmentation. We then align and match characters in a flexible sized gallery with the segmented regions, using parallelised normalised cross-correlation. The experimental results over the Pattern Recognition & Image Analysis Research Lab-Natural History Museum (PRImA-NHM) dataset show remarkably high robustness of the algorithm in classifying cluttered, occluded and noisy samples, in addition to those with significant high missing data. The algorithm, which gives 84.0% classification rate with false positive rate 0.16 over the dataset, does not require training samples and generates compelling results as opposed to the training-based approaches, which have used the same benchmark.

KW - classification

KW - handwritten

KW - machine-printed

KW - OCR

KW - shape analysis

KW - template matching

UR - http://www.scopus.com/inward/record.url?scp=84979523443&partnerID=8YFLogxK

UR - http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=7490151

U2 - 10.1109/DAS.2016.22

DO - 10.1109/DAS.2016.22

M3 - Conference contribution

SN - 978-1-5090-1792-8

SP - 399

EP - 404

BT - Proceedings : 12th IAPR International Workshop on Document Analysis Systems, DAS 2016

PB - IEEE

ER -

Emambakhsh M, He Y, Nabney I. Handwritten and machine-printed text discrimination using a template matching approach. In Proceedings : 12th IAPR International Workshop on Document Analysis Systems, DAS 2016. IEEE. 2016. p. 399-404 https://doi.org/10.1109/DAS.2016.22