Identification of Machine-Printed and Handwritten Words in Arabic and Latin Scripts

Saïdani, Asma; Echi, Afef Kacem; Belaïd, Abdel

doi:10.1109/icdar.2013.163

Cited by 20 publications

(6 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As noted in a previous work [1], [2], Arabic script has lower tbd than Latin because Arabic script has lower discontinuity since it is straighter and flat and does not have high links between letters like in handwritten Latin script (especially in case of o, and v letters). In fact, the high links increase the differences between coordinates of lower profile pixels of word.…”

Section: Features Proposed In the Literaturementioning

confidence: 65%

How to separate between Machine-Printed/Handwritten and Arabic/Latin Words?

Kacem¹,

Saïdani²,

Belaïd³

2014

ELCVIA

View full text Add to dashboard Cite

This paper gathers some contributions to script and its nature identification. Different sets of features have been employed successfully for discriminating between handwritten and machine-printed Arabic and Latin scripts. They include some well established features, previously used in the literature, and new structural features which are intrinsic to Arabic and Latin scripts. The performance of such features is studied towards this paper. We also compared the performance of five classifiers: Bayes (AODEsr), k-Nearest Neighbor (k-NN), Decision Tree (J48), Support Vector Machine (SVM) and Multilayer perceptron (MLP) used to identify the script at word level. These classifiers have been chosen enough different to test the feature contributions. Experiments have been conducted with handwritten and machine-printed words, covering a wide range of fonts. Experimental results show the capability of the proposed features to capture differences between scripts and the effectiveness of the three classifiers. An average identification precision and recall rates of 98.72% was achieved, using a set of 58 features and AODEsr classifier, which is slightly better than those reported in similar works.

show abstract

Section: Features Proposed In the Literaturementioning

confidence: 65%

How to separate between Machine-Printed/Handwritten and Arabic/Latin Words?

Kacem¹,

Saïdani²,

Belaïd³

2014

ELCVIA

View full text Add to dashboard Cite

show abstract

“…Also, in one word, two consecutive PAWs may overlap. For more details about Arabic/Latin script properties, see previous works in [1] and [2]. Table I summarizes the surveyed methods.…”

Section: Arabic and Latin Script Characteristicsmentioning

confidence: 99%

Proposition to distinguish machine-printed from handwritten Arabic and Latin words

Saïdani

Echi

Belaïd

2014

2014 Information and Communication Technologies Innovation and Application (ICTIA)

View full text Add to dashboard Cite

In this work, we gathered some contributions to identify script and its nature. We successfully employed many features to distinguish between handwritten and machine-printed Arabic and Latin scripts at word level. Some of them are previously used in the literature, and the others are here proposed. The new proposed structural features are intrinsic to Arabic and Latin scripts. The performance of all extracted features is studied towards this paper. We also compared the performance of three classifiers: Bayes (AODEsr), k-Nearest Neighbor (k-NN) and Decision Tree (J48), used to identify the script at word level. These classifiers have been chosen enough different to test the feature contributions. We carried experiments using standard databases. Obtained results demonstrate used feature capability to capture differences between scripts. Using a set of 58 selected features and a Bayesbased classifier, we achieved an average identification rate equals to 98.72%, which considered a very satisfactory rate compared to some related works.

show abstract

“…It is a task of recognizing the language in a given document. It is the basic preprocessing step in Optical Character recognition [2].One of the most important and challenging field in pattern recognition is OCR. Optical Character recognition is a process by which scanned document/document is optically scanned, normally by using a flatbed scanner, which is then converted to an electronic format retaining the text script [3].…”

Section: Introductionmentioning

confidence: 99%

Segmentation techniques for handwritten script recognition system

Mathew¹,

Shinde²,

Patil³

2015

2015 International Conference on Circuits, Power and Computing Technologies [ICCPCT-2015]

View full text Add to dashboard Cite

Different scripts are used for writing purposes. It is very much important to identify/know these scripts before an appropriate character recognition algorithm is chosen. This paper deals mainly on the various segmentation techniques available for handwritten script recognition system. It is an important technique for image processing. Image segmentation provides meaningful objects of the image. Identification of a script in handwritten recognition system is really a challenging task in real world. Many researchers are indulged in the recognition of scripts from handwritten documents over four decades. This task becomes more and more challenging when it comes to multi lingual/script country like India. An input Image is preprocessed through various steps and segmented by different available techniques. This identification of script helps in optical Character Recognition systems (OCR)

show abstract

Identification of Machine-Printed and Handwritten Words in Arabic and Latin Scripts

Cited by 20 publications

References 6 publications

How to separate between Machine-Printed/Handwritten and Arabic/Latin Words?

How to separate between Machine-Printed/Handwritten and Arabic/Latin Words?

Proposition to distinguish machine-printed from handwritten Arabic and Latin words

Segmentation techniques for handwritten script recognition system

Contact Info

Product

Resources

About