A line-based representation for matching words in historical manuscripts

Can, Ethem F.; Duygulu, Pınar

doi:10.1016/j.patrec.2011.02.013

Cited by 25 publications

(15 citation statements)

References 62 publications

(70 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In [54], an application of approximate contour lines to text recognition is presented. In this section, we discuss contour line approximation as another baseline approach for Kufic recognition problem.…”

Section: Approximate Contour Line Matchingmentioning

confidence: 99%

“…Can et al [54] exploited Douglas-Peucker algorithm to describe words in handwritten documents as a set of lines. A line is described by its position, orientation, and length as in [53].…”

Section: Approximate Contour Line Matchingmentioning

confidence: 99%

“…Moreover, instances of a sub-pattern in different images may be approximated into different number of lines due to the variations in lighting conditions and sizes. Therefore, the method used in [54] is not feasible for matching Kufic patterns.…”

Section: Approximate Contour Line Matchingmentioning

confidence: 99%

“…In the first version, we used the comparison method in [54] as is. Each line segment is described by its length (l), angle (a), and position of midpoint relative to upper left corner of the component (r).…”

Section: Approximate Contour Line Matchingmentioning

confidence: 99%

“…Firstly, contours of a component are extracted, and contour points on these contours are approximated to lines using the Douglas-Peucker line approximation algorithm [56] similar to [54] (see Fig. 14).…”

Section: Sequence Matching Based On Contour Representationmentioning

confidence: 99%

See 4 more Smart Citations

Matching Islamic patterns in Kufic images

Arifoǧlu

Sahin

Adiguzel

et al. 2015

Pattern Anal Applic

View full text Add to dashboard Cite

In this study, we address the problem of matching patterns in Kufic calligraphy images. Being used as a decorative element, Kufic images have been designed in a way that makes it difficult to be read by non-experts. Therefore, available methods for handwriting recognition are not easily applicable to the recognition of Kufic patterns. In this study, we propose two new methods for Kufic pattern matching. The first method approximates the contours of connected components into lines and then utilizes chain code representation. Sequence matching techniques with a penalty for gaps are exploited for handling the variations between different instances of sub-patterns. In the second method, skeletons of connected components are represented as a graph where junction and end points are considered as nodes. Graph isomorphism techniques are then relaxed for partial graph matching. Methods are evaluated over a collection of 270 square Kufic images with 8,941 sub-patterns. Experimental results indicate that, besides retrieval and indexing of known patterns, our method also allows the discovery of new patterns.

show abstract

Section: Approximate Contour Line Matchingmentioning

confidence: 99%

“…Can et al [54] exploited Douglas-Peucker algorithm to describe words in handwritten documents as a set of lines. A line is described by its position, orientation, and length as in [53].…”

Section: Approximate Contour Line Matchingmentioning

confidence: 99%

Section: Approximate Contour Line Matchingmentioning

confidence: 99%

Section: Approximate Contour Line Matchingmentioning

confidence: 99%

Section: Sequence Matching Based On Contour Representationmentioning

confidence: 99%

See 3 more Smart Citations

Matching Islamic patterns in Kufic images

Arifoǧlu

Sahin

Adiguzel

et al. 2015

Pattern Anal Applic

View full text Add to dashboard Cite

show abstract

A deep learning model for Ottoman OCR

DÖLEK>

Kurt

2022

Concurrency and Computation

View full text Add to dashboard Cite

The Ottoman OCR is an open problem because the OCR models for Arabic do not perform well on Ottoman. The models specifically trained with Ottoman documents have not produced satisfactory results either. We present a deep learning model and an OCR tool using that model for the OCR of printed Ottoman documents in the naksh font. We propose an end-to-end trainable CRNN architecture consisting of CNN, RNN (LSTM), and CTC layers for the Ottoman OCR problem. An experimental comparison of this model, called Osmanlica.com, with the Tesseract Arabic, the Tesseract Persian, Abby Finereader, Miletos, and Google Docs OCR tools or models was performed using a test data set of 21 pages of original documents. With 88.86% raw text, 96.12% normalized text, and 97.37% joined text character recognition accuracy, the Osmanlica.com Hybrid model outperforms the others with a marked difference. Our model outperforms the next best model by a clear margin of 4% which is a significant improvement considering the difficulty of the Ottoman OCR problem, and the huge size of the Ottoman archives to be processed. The hybrid model also achieves 58% word recognition accuracy on normalized text which is the only rate above 50%.

show abstract

Searching and Browsing in Historical Documents—State of the Art and Novel Approaches for Template-Based Keyword Spotting

Stauffer

Fischer

Riesen

2018

Business Information Systems and Technology 4.0

View full text Add to dashboard Cite

In many public and private institutions, the digitalization of handwritten 1 documents has progressed greatly in recent decades. As a consequence, the number 2 of handwritten documents that are available digitally is constantly increasing. How-3 ever, accessibility to these documents in terms of browsing and searching is still an 4 issue as automatic full transcriptions are often not feasible. To bridge this gap, Key-5 word Spotting (KWS) has been proposed as a flexible and error-tolerant alternative 6 to full transcriptions. KWS provides unconstrained retrievals of keywords in hand-7 written documents that are acquired either online or offline. In general, offline KWS 8 is regarded as the more difficult task when compared to online KWS where temporal 9 information on the writing process is also available. The focus of this chapter is on 10 handwritten historical documents and thus on offline KWS. In particular, we review 11 and compare different state-of-the-art as well as novel approaches for template-based 12 KWS. In contrast to learning-based KWS, template-based KWS can be applied to 13 documents without any a priori learning of a model and is thus regarded as the more 14 flexible approach. 15

show abstract

A line-based representation for matching words in historical manuscripts

Cited by 25 publications

References 62 publications

Matching Islamic patterns in Kufic images

Matching Islamic patterns in Kufic images

A deep learning model for Ottoman OCR

Searching and Browsing in Historical Documents—State of the Art and Novel Approaches for Template-Based Keyword Spotting

Contact Info

Product

Resources

About