Labeling, Cutting, Grouping: An Efficient Text Line Segmentation Method for Medieval Manuscripts

Alberti, MicheleDocument Image and Voice Analysis Group (DIVA), University of Fribourg, Switzerland (author)

Labeling, Cutting, Grouping : An Efficient Text Line Segmentation Method for Medieval Manuscripts

ISBN för värdpublikation: 978-1-7281-3014-9, 978-1-7281-3015-6
This paper introduces a new way for text-line extraction by integrating deep-learning based pre-classification and state-of-the-art segmentation methods. Text-line extraction in complex handwritten documents poses a significant challenge, even to the most modern computer vision algorithms. Historical manuscripts are a particularly hard class of documents as they present several forms of noise, such as degradation, bleed-through, interlinear glosses, and elaborated scripts. In this work, we propose a novel method which uses semantic segmentation at pixel level as intermediate task, followed by a text-line extraction step. We measured the performance of our method on a recent dataset of challenging medieval manuscripts and surpassed state-of-the-art results by reducing the error by 80.7%. Furthermore, we demonstrate the effectiveness of our approach on various other datasets written in different scripts. Hence, our contribution is two-fold. First, we demonstrate that semantic pixel segmentation can be used as strong denoising pre-processing step before performing text line extraction. Second, we introduce a novel, simple and robust algorithm that leverages the high-quality semantic segmentation to achieve a text-line extraction performance of 99.42% line IU on a challenging dataset.

Vögtlin, LarsDocument Image and Voice Analysis Group (DIVA), University of Fribourg, Switzerland (author)
Pondenkandath, VinaychandranDocument Image and Voice Analysis Group (DIVA), University of Fribourg, Switzerland (author)
Seuret, MathiasDocument Image and Voice Analysis Group (DIVA), University of Fribourg, Switzerland. Pattern Recognition Lab, Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany (author)
Ingold, RolfDocument Image and Voice Analysis Group (DIVA), University of Fribourg, Switzerland (author)
Liwicki, MarcusLuleå tekniska universitet,EISLAB,Document Image and Voice Analysis Group (DIVA), University of Fribourg, Switzerland(Swepub:ltu)marliw (author)
Document Image and Voice Analysis Group (DIVA), University of Fribourg, SwitzerlandDocument Image and Voice Analysis Group (DIVA), University of Fribourg, Switzerland. Pattern Recognition Lab, Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany (creator_code:org_t)

In:The 15th IAPR International Conference on Document Analysis and Recognition: IEEE, s. 1200-1206

By the author/editor: Alberti, Michele; Vögtlin, Lars; Pondenkandath, V ...; Seuret, Mathias; Ingold, Rolf; Liwicki, Marcus

About the subject

NATURAL SCIENCES: NATURAL SCIENCES; and Computer and Inf ...; and Computer Science ...

Kungliga biblioteket hanterar dina personuppgifter i enlighet med EU:s dataskyddsförordning (2018), GDPR. Läs mer om hur det funkar här.
Så här hanterar KB dina uppgifter vid användning av denna tjänst.