New Binarization Approach Based on Text Block Extraction

Messaoud, Ines Ben; Amiri, Hamid; Abed, Haikal El; Märgner, Volker

doi:10.1109/icdar.2011.243

Cited by 22 publications

(6 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Moghaddam et al [26] estimate the backdrop surface of the document by an adaptive and iterative image averaging approach. Messaoud et al [27] apply a binarization technique to selected items of interest by combining a preprocessing stage and a localization step. Pardhi et al [28] construct local thresholds by a combination of local image contrast and gradient combination to segment text and it also an adaptive image contrast technique.…”

Section: Mixed Threshold Methodsmentioning

confidence: 99%

Summary of Document Image Binarization

Yang,

Zhang,

Zhou

et al. 2024

Preprint

View full text Add to dashboard Cite

Document image binarization is a challenging task, especially when it comes to text segmentation in degraded document images. The binarization, as a pre-processing step of optical character recognition (OCR), is one of the most fundamental and commonly used segmentation methods. It separates the foreground text from the background of the document image to facilitate subsequent image processing. In view of the different degradation degree of document image, researchers have proposed a variety of solutions. This paper reviews the main binarization techniques, including both traditional algorithms and deep learning-based algorithms. We also summarize some difficulties and challenges in the field of document image binarization. Here, we evaluate various image binarization techniques to identify shortcomings in current methods and provide some help for future research.

show abstract

Section: Mixed Threshold Methodsmentioning

confidence: 99%

Summary of Document Image Binarization

Yang,

Zhang,

Zhou

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…P SNR = 10 · log10 α 2 MSE (15) α is the difference between background and foreground pixel intensities α = 1, MSE is the mean square error …”

Section: Evaluation Metricsmentioning

confidence: 99%

“…The set M is composed of L = 6 different binarization methods, one is a global binarization, the well known Otsu's method [10], where only one threshold for the entire image is used. The other methods are considered as local bianrization, Bernsen [11], Niblack [12], Sauvola [13], Gatos [14] and Ben Messaoud [15], where for each pixel in the input image a threshold is returned according to the intensities of the neighborhood pixels or pixel intensities belonging to a specific window. Gatos and Ben Messaoud's methods are combined with the Wiener's filter.…”

Section: Selection Phasementioning

confidence: 99%

A design of a preprocessing framework for large database of historical documents

Messaoud

Abed

Märgner

et al. 2011

Proceedings of the 2011 Workshop on Historical Document Imaging and Processing

Self Cite

View full text Add to dashboard Cite

The objective of document preprocessing is to ease the text recognition or the document indexing processes. The analysis of historical documents seems to be a big challenge because the majority of those documents are noisy and present many degradations. In this paper we propose a preprocessing framework for a large dataset of historical documents. The proposed framework is decomposed of two phases, the selection and the evaluation. During the first phase one or multiple methods are corresponded for each book of the used database. The validation of the selection results is performed during the evaluation. The experiments are applied on printed and handwritten documents extracted respectively from Google-Books and Bayerische Staatsbibliothek databases. The results returned during the evaluation are very promising.

show abstract

“…In local adaptive techniques [3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19][20][21][22], a threshold value is determined for each pixel depending on the neighboring pixels within a local window. Proper choice of threshold value leads to the high quality of binary image.…”

Section: Introductionmentioning

confidence: 99%

Local Adaptive Automatic Binarisation (LAAB)

RomenSingh¹,

Roy²,

Singh³

2012

IJCA

View full text Add to dashboard Cite

Most of the binarization techniques associate a certain intensity value called threshold which separate the pixel values of the concerned input grey scale image into two classes like background and foreground. Each and every pixel should be compared with the threshold and transformed to its respective class according to the threshold value. In this paper an automatic binarisation technique with local adaptation without any intensity value (threshold) of partition, is described. It creates a binarised image by transforming the input image to its respective binarised image automatically without using any threshold value. It uses local mean to adapt to local environment within a window of size w w. Local mean determination is time consuming one and to reduce the time consumption, integral sum image is used as prior process. The input grey scale image is self transformed to an integral sum image within itself and then transform to binary image from the integral sum image itself.

show abstract

New Binarization Approach Based on Text Block Extraction

Cited by 22 publications

References 10 publications

Summary of Document Image Binarization

Summary of Document Image Binarization

A design of a preprocessing framework for large database of historical documents

Local Adaptive Automatic Binarisation (LAAB)

Contact Info

Product

Resources

About