Recognizing handwritten Arabic words using grapheme segmentation and recurrent neural networks

Abandah, Gheith A.; Jamour, Fuad; Qaralleh, Esam A. Al

doi:10.1007/s10032-014-0218-7

Cited by 40 publications

(42 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Sub-word segmentation is addressed as an intermediate stage of a system meant to recognize handwritten Arabic words in [9]. Firstly, the input word is segmented into sub-words.…”

Section: Comparative Resultsmentioning

confidence: 99%

“…C1={(5,10), (4,9), (5,9), (6,10), (7,9), (7,8), (8,9), (7,7), (5,7), (5,8), (6,6), (7,6), (4,5), (2,5), (3,5)} , C2={(2,7), (3,7)} , C3={(9,7), (10,6), (10,5), (9,4), (8,4), (8,3)} and C4={(6,4), (6,3), (6,2).…”

Section: Methodsunclassified

“…Finally, each primary component along with the associated secondary components are marked as PAW and passed to the subsequent stages of character segmentation and recognition. The technique used for sub-word segmentation in [9] proceeds as follows: word baseline is estimated using the horizontal projection histogram method. Then, the main and secondary bodies are identified; main bodies of the subwords are extracted and the secondary bodies are assigned using predefined rules to their respective main bodies to yield the sub-word.…”

Section: Related Workmentioning

confidence: 99%

“…The components C3={(9,7), (10,6), (10,5), (9,4), (8,4), (8,3)} and C4=={(6,4), (6,3), (6,2) of image given in Fig. 4 is reconstructed as follows-firstly, the (xmin,xmax) and (ymin,ymax) of each component is retrieved ((xmin,xmax) and (ymin,ymax) of C3 are (8,10) and (3,7) respectively.…”

Section: Algorithm: Reconstruct Sub-imagementioning

confidence: 99%

See 3 more Smart Citations

Graph Modeling based Segmentation of Handwritten Arabic Text into Constituent Subwords

Ghaleb¹,

Nagabhushan²,

Pal³

2016

IJIGSP

View full text Add to dashboard Cite

Abstract-Segmentation of Arabic text is a major challenge that shall be addressed by any recognition system. The cursive nature of Arabic writing makes it necessary to handle the segmentation issue at various levels. Arabic text line can be viewed as a sequence of words which in turn can be viewed as a sequence of subwords. Sub-words have the frequently encountered intrinsic property of sharing the same vertical space which makes vertical projection based segmentation technique inefficient. In this paper, the task of segmenting handwritten Arabic text at sub-word level is taken up. The proposed algorithm is based on pulling away the connected components to overcome the impossibility of separating them by vertical projection based approach. Graph theoretic modeling is proposed to solve the problem of connected component extraction. In the sequel, these components are subjected to thorough analysis in order to obtain the constituent sub-words where a sub-word may consist of many components. The proposed algorithm was tested using variety of handwritten Arabic samples taken from different databases and the results obtained are encouraging.

show abstract

“…Sub-word segmentation is addressed as an intermediate stage of a system meant to recognize handwritten Arabic words in [9]. Firstly, the input word is segmented into sub-words.…”

Section: Comparative Resultsmentioning

confidence: 99%

Section: Methodsunclassified

Section: Related Workmentioning

confidence: 99%

Section: Algorithm: Reconstruct Sub-imagementioning

confidence: 99%

See 2 more Smart Citations

Graph Modeling based Segmentation of Handwritten Arabic Text into Constituent Subwords

Ghaleb¹,

Nagabhushan²,

Pal³

2016

IJIGSP

View full text Add to dashboard Cite

show abstract

“…Recognition based on segment ligatures and open characters including segmentation region has given accuracy results of over 98% recognition rate as displayed in Tables 2 which has been tested for different handwriting datasets IFN/ENITdatabase Arabic OCR handwritten [14] with 2% oversegmentation. The measurement segmentation performance tests for each experiment test targets the contained open Arabic letters ‫س,ي(‬ ‫)ْ.ة.ق‬ and the results are presented in Table 2.…”

Section: Evaluation and Analysismentioning

confidence: 99%

Efficient Approach to Segment Ligatures and Open Characters in offline Arabic Text

2017

IJCCIE

View full text Add to dashboard Cite

Abstract-This paper researches offline Arabic handwriting recognition. It introduces a new approach to segmentation ligature and open Arabic character based on the structural perspective dealing with sub-words/words, including dots to recognize individual letters. Segmentation approaches that have been integrated into the recognition phase have the capability to deal with ligatures and closed characters issues. This complex problem is due to the cursive writing nature of the Arabic language. This paper also develops an Arabic character algorithm on segmented, pixel based and centre reign (CR) to recognize the letters. The evaluation results are stated in the IFN/ENIT and IAM database which indicate the recognition rate and the effectiveness of our system.

show abstract

Improving Handwritten Arabic Text Recognition Using an Adaptive Data-Augmentation Algorithm

Eltay

Zidouri

Ahmad

et al. 2021

Document Analysis and Recognition – ICDAR 2021 Workshops

View full text Add to dashboard Cite

Recognizing handwritten Arabic words using grapheme segmentation and recurrent neural networks

Cited by 40 publications

References 40 publications

Graph Modeling based Segmentation of Handwritten Arabic Text into Constituent Subwords

Graph Modeling based Segmentation of Handwritten Arabic Text into Constituent Subwords

Efficient Approach to Segment Ligatures and Open Characters in offline Arabic Text

Improving Handwritten Arabic Text Recognition Using an Adaptive Data-Augmentation Algorithm

Contact Info

Product

Resources

About