Beyond the ground truth: Alternative quality measures of document binarizations

Arie Shaus, Barak Sober, Eli Turkel, Eli Piasetzky

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

6 Scopus citations

Abstract

This article discusses the quality assessment of binary images. The customary, ground truth based methodology, used in the literature is shown to be problematic due to its subjective nature. Several previously suggested alternatives are surveyed and are also found to be inadequate in certain scenarios. A new approach, quantifying the adherence of a binarization to its document image is proposed and tested using six different measures of accuracy. The measures are evaluated experimentally based on datasets from DIBCO and H-DIBCO competitions, with respect to different kinds of binarization degradations.

Original languageEnglish
Title of host publicationProceedings - 2016 15th International Conference on Frontiers in Handwriting Recognition, ICFHR 2016
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages495-500
Number of pages6
ISBN (Electronic)9781509009817
DOIs
StatePublished - 2 Jul 2016
Event15th International Conference on Frontiers in Handwriting Recognition, ICFHR 2016 - Shenzhen, China
Duration: 23 Oct 201626 Oct 2016

Publication series

NameProceedings of International Conference on Frontiers in Handwriting Recognition, ICFHR
Volume0
ISSN (Print)2167-6445
ISSN (Electronic)2167-6453

Conference

Conference15th International Conference on Frontiers in Handwriting Recognition, ICFHR 2016
Country/TerritoryChina
CityShenzhen
Period23/10/1626/10/16

Funding

FundersFunder number
Seventh Framework Programme
European Research Council229418
Israel Science Foundation644/08, 1457/13
Tel Aviv University
Seventh Framework Programme

    Keywords

    • Binarization
    • Evaluation
    • Ground truth
    • Quality measure

    Fingerprint

    Dive into the research topics of 'Beyond the ground truth: Alternative quality measures of document binarizations'. Together they form a unique fingerprint.

    Cite this