A Hybrid for Line Segmentation in Handwritten Documents

Adiguzel, Hande; Sahin, Emre; Duygulu, Pınar

doi:10.1109/icfhr.2012.156

Cited by 18 publications

(10 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The algorithm they used was based on the seam carving approach, which is mostly used for content aware image resizing. Adiguzel et al [11] presented in 2012, an approach for text line segmentation which combines connected component based and projection based information to take advantage of aspects of both methods. Palakollu et al proposed [12], a language specific property of header and base line for segmentation of Devanagari Text.…”

Section: Literature Surveymentioning

confidence: 99%

Line Segmentation in Handwritten Assamese and Meetei Mayek Script Using Seam Carving Based Algorithm

Kumar

Kalita

2015

Springer Proceedings in Physics

View full text Add to dashboard Cite

Line segmentation is a key stage in an Optical Character Recognition system. This paper primarily concerns the problem of text line extraction on color and grayscale manuscript pages of two major North-east Indian regional Scripts, Assamese and Meetei Mayek. Line segmentation of handwritten text in Assamese and Meetei Mayek scripts is an uphill task primarily because of the structural features of both the scripts and varied writing styles. Line segmentation of a document image is been achieved by using the Seam carving technique, in this paper. Researchers from various regions used this approach for content aware resizing of an image. However currently many researchers are implementing Seam Carving for line segmentation phase of OCR. Although it is a language independent technique, mostly experiments are done over Arabic, Greek, German and Chinese scripts. Two types of seams are generated, medial seams approximate the orientation of each text line, and separating seams separated one line of text from another. Experiments are performed extensively over various types of documents and detailed analysis of the evaluations reflects that the algorithm performs well for even documents with multiple scripts. In this paper, we present a comparative study of accuracy of this method over different types of data.

show abstract

Section: Literature Surveymentioning

confidence: 99%

Line Segmentation in Handwritten Assamese and Meetei Mayek Script Using Seam Carving Based Algorithm

Kumar

Kalita

2015

Springer Proceedings in Physics

View full text Add to dashboard Cite

show abstract

“…In [19], the data set consist of images of handwritten Arabic containing each two pages, after removing small components and linking broken characters, Fourier curve fitting within the horizontal PP is used to locate the point of separation. The contour is used to extract the base line of the CC which allows locating the cut point between different adjacent lines.…”

Section: Related Workmentioning

confidence: 99%

“…The methods which are based on CC analysis use geometric information such as shape, orientation, position, and size of CC for grouping them in rows. These methods are more suitable for complex documents than methods based on PP, even if they could be sensitive to changes in the component structures [19].…”

Section: Proposed Approachmentioning

confidence: 99%

Arabic handwritten text line extraction using connected component analysis from a multi agent perspective

Boulid

Souhar

Elkettani

2015

2015 15th International Conference on Intelligent Systems Design and Applications (ISDA)

View full text Add to dashboard Cite

This paper presents an approach to extract curved text lines from Arabic handwritten documents, based on the perception mechanisms involved in the human reading process. Our approach is based on multi-agent systems to detect and group connected components that belong to the same line. This proposed system makes use of information about features and arrangement of those components. Experimental results on a data-set of Arabic handwritten documents show that this approach is a promising solution for extracting handwritten curved text lines.

show abstract

“…These methods are more suitable for complex documents compared to methods based on Projection Profile (PP) [9,10], even if they could be sensitive to changes in the size and component structures [11].…”

Section: Proposed Approachmentioning

confidence: 99%

Detection of Text Lines of Handwritten Arabic Manuscripts using Markov Decision Processes

Boulid¹,

Souhar²,

Elkettani³

2016

IJIMAI

View full text Add to dashboard Cite

-In a character recognition systems, the segmentation phase is critical since the accuracy of the recognition depend strongly on it. In this paper we present an approach based on Markov Decision Processes to extract text lines from binary images of Arabic handwritten documents. The proposed approach detects the connected components belonging to the same line by making use of knowledge about features and arrangement of those components. The initial results show that the system is promising for extracting Arabic handwritten lines.

show abstract

A Hybrid for Line Segmentation in Handwritten Documents

Cited by 18 publications

References 17 publications

Line Segmentation in Handwritten Assamese and Meetei Mayek Script Using Seam Carving Based Algorithm

Line Segmentation in Handwritten Assamese and Meetei Mayek Script Using Seam Carving Based Algorithm

Arabic handwritten text line extraction using connected component analysis from a multi agent perspective

Detection of Text Lines of Handwritten Arabic Manuscripts using Markov Decision Processes

Contact Info

Product

Resources

About