Machine recognition of printed Arabic text utilizing natural language morphology

Amin, Adnan; Al-Fedaghi, Sabah

doi:10.1016/s0020-7373(05)80160-9

Cited by 34 publications

(8 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Segmentation of Farsi words is base on histogram and contour [18][19][20][21]. The steps of character segmentation are as follows:…”

Section: Farsi Character Segmentationmentioning

confidence: 99%

“…Step 2: Word segmentation: The Farsi words in each line of text are separated by vertical projection and connected comments [18][19][20][21].…”

Section: Farsi Character Segmentationmentioning

confidence: 99%

“…Step 4: Character segmentation: In this step we segmented cursive words or sub words into their building characters [21]. For example word ( ) has two isolated letters ( ) and a cursive sub word ( ) with four cursive characters.…”

Section: Farsi Character Segmentationmentioning

confidence: 99%

See 2 more Smart Citations

Fast Zernike wavelet moments for Farsi character recognition

Broumandnia¹,

Shanbehzadeh²

2007

Image and Vision Computing

View full text Add to dashboard Cite

“…Segmentation of Farsi words is base on histogram and contour [18][19][20][21]. The steps of character segmentation are as follows:…”

Section: Farsi Character Segmentationmentioning

confidence: 99%

“…Step 2: Word segmentation: The Farsi words in each line of text are separated by vertical projection and connected comments [18][19][20][21].…”

Section: Farsi Character Segmentationmentioning

confidence: 99%

See 1 more Smart Citation

Fast Zernike wavelet moments for Farsi character recognition

Broumandnia¹,

Shanbehzadeh²

2007

Image and Vision Computing

View full text Add to dashboard Cite

“…Examples of this approach are in [7,14,26,28,30]. The difficulty of segmenting the symbols of a word leads to inaccuracies in the segmentation and, as a result, in the recognition.…”

Section: Motivationmentioning

confidence: 99%

A segmentation-free approach to text recognition with application to Arabic text

Al-Badr

1998

International Journal on Document Analysis and Recognition

View full text Add to dashboard Cite

Abstract. In recognizing cursive scripts, a major undertaking is segmenting cursive words into characters and isolating merged characters. The segmentation is usually the pivotal stage in the system to which a sizable portion of processing is devoted and a considerable share of recognition errors is attributed. The most notable feature of Arabic writing is its cursiveness. Compared to other features, the cursiveness of Arabic words poses the most difficult problem for recognition algorithms. In this work, we describe the design and implementation of an Arabic word recognition system. To recognize a word, the system does not segment it into characters in advance; rather, it recognizes the input word by detecting a set of "shape primitives" on the word. It then matches the regions of the word (represented by the detected primitives) with a set of symbol models. A spatial arrangement of symbol models that are matched to regions of the word, then, becomes the description of the recognized word. Since the number of potential arrangements of all symbol models is combinatorially large, the system imposes a set of constraints that pertain to word structure and spatial consistency. The system searches the space made up of the arrangements that satisfy the constraints, and tries to maximize the a posteriori probability of the arrangement of symbol models. We measure the accuracy of the system not only on words but on isolated characters as well. For isolated characters, it has a recognition rate of 99.7% for synthetically degraded symbols and 94.1% for scanned symbols. For isolated words the system has a recognition rate of 99.4% for noise-free words, 95.6% for synthetically degraded words, and 73% for scanned words.

show abstract

“…Intensive research has been carried out in this area with a large number of technical papers and reports in the literature devoted to character recognition. This subject has attracted immense research interest not only because of the very challenging nature of the problem, but also because it provides the means for automatic processing of large volumes of data in postal code reading [I, 2], office automation [3,4], and other business and scientific applications [5][6][7].…”

Section: Introductionmentioning

confidence: 99%

Recognition of Printed Arabic Text via Machine Learning

Amin¹

1999

International Conference on Advances in Pattern Recognition

Self Cite

View full text Add to dashboard Cite

The main theme of this paper is the automatic recognition of Arabic printed text using machine learning C4.5. The technique can be divided into three major steps: The first step is pre-processing in which the original image is transformed into a binary image utilizing a 300 dpi scanner and then forming the connected component. At the moment we extract seven types of global features: number of subwords, number of peaks of each subwords, number of loops of each peak, number and position of complementary characters,. Second, global features of the input Arabic word are then extracted such as number subwords, number of peaks within the subword, number of loops of each peak, number the height and width of each peak and position of the complementary character, etc .. Finally, Machine learning C4.) is used for character classification to generate a decision tree.The system was tested with a set of 500 different classes of Arabic words (each class has fifteen different fonts) using Cross validation algorithm and the average recognition rate obtained was 96.52%. This is a very promising result and clearly shows that machine learning techniques are well suited to these types of application.Many papers have been concerned with the recognition of Latin, Chinese and Japanese characters. However, although almost a third of a billion people worldwide, in several different languages, use Arabic characters for writing, little research progress, in both on-line and off-line has been achieved towards the automatic recognition of Arabic characters. This is a result of the lack of adequate support in term of funding, and other utilities such as Arabic text database, dictionaries, etc .. and of course of the cursive nature of its writing rules. Surveys on Arabic characters recognition can be found in [8][9][10][11].This work describes the design and implementation of a new technique to recognize Arabic words. To recognize an input word, the system does not segment the word in advance S. Singh (ed.), International Conference on Advances in Pattern Recognition

show abstract

Machine recognition of printed Arabic text utilizing natural language morphology

Cited by 34 publications

References 6 publications

Fast Zernike wavelet moments for Farsi character recognition

Fast Zernike wavelet moments for Farsi character recognition

A segmentation-free approach to text recognition with application to Arabic text

Recognition of Printed Arabic Text via Machine Learning

Contact Info

Product

Resources

About