READ-BAD: A New Dataset and Evaluation Scheme for Baseline Detection in Archival Documents

In recent years there have been multiple successful attempts tackling document processing problems separately by designing task specific hand-tuned strategies. We argue that the diversity of historical document processing tasks prohibits to solve them one at a time and shows a need for designing generic approaches in order to handle the variability of historical series. In this paper, we address multiple tasks simultaneously such as page extraction, baseline extraction, layout analysis or multiple typologies of illustrations and photograph extraction.We propose an open-source implementation of a CNN-based pixel-wise predictor coupled with task dependent post-processing blocks. We show that a single CNN-architecture can be used across tasks with competitive results. Moreover most of the task-specific post-precessing steps can be decomposed in a small number of simple and standard reusable operations, adding to the flexibility of our approach.

show abstract

“…most characters rest upon and descenders extend below". The READ-BAD dataset [22] has been used for the cBAD: ICDAR2017 Competition [8].…”

Section: B Baseline Detectionmentioning

confidence: 99%

dhSegment: A Generic Deep-Learning Approach for Document Segmentation

Oliveira

Séguin

Kaplan

2018

2018 16th International Conference on Frontiers in Handwriting Recognition (ICFHR)

149

143

View full text Add to dashboard Cite

show abstract

“…There are many possibilities to determine if two text lines are close to each other or not. Here, the well-established method of [3] is used. We say two text lines are close, if their baselines are geometrically close to each other (see Section III-B for details).…”

Section: B Using Geometric Information As Restrictionmentioning

confidence: 99%

“…To define the neighborhood of G x we use a method that compares the so-called baselines of the text lines. This is a common measure to evaluate the performance of a layout analysis result [3]. We call the tuple of twodimensional points B "`B 1 , .…”

Section: B Restricting By Geometric Positionmentioning

confidence: 99%

“…which implicitly restricts the set of valid assignment matrices in (3). Indeed, setting baselines to be close if they have any connection is probably a very soft restriction, but reasonable to avoid erroneously non-assignments for close H y and G x .…”

Section: B Restricting By Geometric Positionmentioning

confidence: 99%

“…For the LA, which extracts text lines on pixel level, there are evaluation schemes based on different entities. For instance, based on pixel information [2], baselines [3] or origin points [4]. Each of these schemes has its application area and consequently its right to exist.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

End-to-End Measure for Text Recognition

Leifert

Labahn

Grüning³

et al. 2019

2019 International Conference on Document Analysis and Recognition (ICDAR)

Self Cite

View full text Add to dashboard Cite

Measuring the performance of text recognition and text line detection engines is an important step to objectively compare systems and their configuration. There exist wellestablished measures for both tasks separately. However, there is no sophisticated evaluation scheme to measure the quality of a combined text line detection and text recognition system. The F-measure on word level is a well-known methodology, which is sometimes used in this context. Nevertheless, it does not take into account the alignment of hypothesis and ground truth text and can lead to deceptive results. Since users of automatic information retrieval pipelines in the context of text recognition are mainly interested in the end-to-end performance of a given system, there is a strong need for such a measure. Hence, we present a measure to evaluate the quality of an end-to-end text recognition system. The basis for this measure is the well established and widely used character error rate, which is limited -in its original form -to aligned hypothesis and ground truth texts. The proposed measure is flexible in a way that it can be configured to penalize different reading orders between the hypothesis and ground truth and can take into account the geometric position of the text lines. Additionally, it can ignore over-and under-segmentation of text lines. With these parameters it is possible to get a measure fitting best to its own needs.Index Terms-measure, end-to-end, character error rate, word error rate, F-measure, bag-of-word, HTR

show abstract

Multi-task Layout Analysis of Handwritten Musical Scores

Quirós

Toselli

Vidal

2019

Pattern Recognition and Image Analysis

View full text Add to dashboard Cite

Document Layout Analysis (DLA) is a process that must be performed before attempting to recognize the content of handwritten musical scores by a modern automatic or semiautomatic system. DLA should provide the segmentation of the document image into semantically useful region types such as staff, lyrics, etc. We present a system that extend the ideas of DLA for handwritten text documents to perform region segmentation, region classification and baseline detection over handwritten musical scores in an integrated manner. Several experiments were carried out on two different datasets in order to validate this approach and assess it in different scenarios. Results show high accuracy in such complex manuscripts and very competent computational time, which is a good indicator of the scalability of the method for very large collections.

show abstract

READ-BAD: A New Dataset and Evaluation Scheme for Baseline Detection in Archival Documents

Cited by 58 publications

References 22 publications

dhSegment: A Generic Deep-Learning Approach for Document Segmentation

dhSegment: A Generic Deep-Learning Approach for Document Segmentation

End-to-End Measure for Text Recognition

Multi-task Layout Analysis of Handwritten Musical Scores

Contact Info

Product

Resources

About