Automatic accurate broken character restoration for patrimonial documents

Allier, Bénédicte; Bali, Nadia; Emptoz, Hubert

doi:10.1007/s10032-005-0012-7

Cited by 22 publications

(5 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another important related research area is that of character enhancement for documents exhibiting various kinds of aging-or printing process-related degradation. In this category fall approaches for character restoration using various energy minimization-and stroke models [5], as well algorithms specialized for the enhancement of lowresolution fax images [6] or typewritten documents [7]. As such methods restrict themselves to strictly improving the visual aspect and connectedness of individual characters they are of limited use in our case.…”

Section: Related Workmentioning

confidence: 99%

Character Enhancement for Historical Newspapers Printed Using Hot Metal Typesetting

Konya

Eickeler

Seibert

2011

2011 International Conference on Document Analysis and Recognition

View full text Add to dashboard Cite

We propose a new method for an effective removal of the printing artifacts occurring in historical newspapers which are caused by problems in the hot metal typesetting, a widely used printing technique in the late 19 th and early 20 th century. Such artifacts typically appear as thin lines between single characters or glyphs and are in most cases connected to one of the neighboring characters. The quality of the optical character recognition (OCR) is heavily influenced by this type of printing artifacts. The proposed method is based on the detection of (near) vertical segments by means of directional single-connected chains (DSCC). In order to allow the robust processing of complex decorative fonts such as Fraktur, a set of rules is introduced. This allows us to successfully process prints exhibiting artifacts with a stroke width even higher than that of most thin characters stems. We evaluate our approach on a dataset consisting of old newspaper excerpts printed using Fraktur fonts. The recognition results on the enhanced images using two independent OCR engines (ABBYY FineReader and Tesseract) show significant improvements over the originals.

show abstract

Section: Related Workmentioning

confidence: 99%

Character Enhancement for Historical Newspapers Printed Using Hot Metal Typesetting

Konya

Eickeler

Seibert

2011

2011 International Conference on Document Analysis and Recognition

View full text Add to dashboard Cite

show abstract

“…In most cases, lossy JPEG compression is chosen. Since the degradations resulting form this compression are irreversible, many people proceed by a restoration phase that consists of a smoothing of the artifacts, sometimes even an attempt to recover the dissolved portions of the texts usually by morphological approaches as used in [16] or by active contours as attempted in [15] and [17], as a result they fall most of the time in a paradigm which is "restore to recognize and recognize to restore". This is why many approaches fail in front of the degradations introduced by excessive JPEG compression.…”

Section: Excessive Compression and Poor Resolutionmentioning

confidence: 99%

Local Orientation Extraction for Wordspotting in Syriac Manuscripts

Bilane

Bres

Emptoz

2008

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. This paper presents a contribution to Word Spotting applied for digitized Syriac manuscripts. The Syriac language was wrongfully accused of being a dead language and has been set aside by the domain of handwriting recognition. Yet it is a very fascinating handwriting that combines the word structure and calligraphy of the Arabic handwriting with the particularity of being intentionally written tilted by an angle of approximately 45°. For the spotting process, we developed a method that should find all occurrences of a certain query word image, based on a selective sliding window technique, from which we extract directional features and afterwards perform a matching using Euclidean distance correspondence between features. The proposed method does not require any prior information, and does not depend of a word to character segmentation algorithm which would be extremely complex to realize due to the tilted nature of the handwriting.

show abstract

“…These in-painting methods do not use character models for the reconstruction and this causes smearing in the in-painted area, and inability to restore severely broken characters. Allier et al [1] restore broken printed Latin characters using the classical active contour model, with a GVF field and a character model. The model is automatically selected using a bank of Gabor filters, and the completion is done in the degraded area.…”

Section: Introductionmentioning

confidence: 99%

Occluded Character Restoration Using Active Contour with Shape Priors

Cohen

Kedem

Dinstein

et al. 2012

2012 International Conference on Frontiers in Handwriting Recognition

View full text Add to dashboard Cite

Broken or partially visible characters is common phenomenon in historical documents. It stems from various factors, such as overlaid text or degradation. Restoring such characters is necessary for document analysis applications. This paper presents a new approach for restoring underlaying Hebrew broken characters that were partially occluded by Arabic text in a palimpsest. We apply text recognition on the fragments of the Hebrew letters and select the k candidate letters that match best the fragments of the Hebrew letters. We then complete the broken Hebrew characters using active contour with the k candidates as shape priors. We use a modified geodesic active contour, which we tailored to occluded text restoration. It is initialized on the fragments of the Hebrew text, then it undergoes an expansion phase and a contraction phase via the occluding Arabic text to form the restored Hebrew character. We measure the distance between the completed character and its corresponding priors and choose the shape with the minimal distance as the reconstructed character. Experimental results are presented. On average 68% of the characters were correctly reconstructed.

show abstract

Automatic accurate broken character restoration for patrimonial documents

Cited by 22 publications

References 25 publications

Character Enhancement for Historical Newspapers Printed Using Hot Metal Typesetting

Character Enhancement for Historical Newspapers Printed Using Hot Metal Typesetting

Local Orientation Extraction for Wordspotting in Syriac Manuscripts

Occluded Character Restoration Using Active Contour with Shape Priors

Contact Info

Product

Resources

About