Automatic segmentation for Arabic characters in handwriting documents

Lawgali, Ahmed; Bouridane, Ahmed; Angelova, Maia; Ghassemlooy, Zabih

doi:10.1109/icip.2011.6116476

Cited by 20 publications

(30 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…These stages are line segmentation, object segmentation, and shape segmentation. A segmentation algorithm was proposed for the computation of the baseline of each subword [8]. The segmentation stage depends on the nature of the Arabic script by combining the information of the writing direction and the characteristics of neighborhood geometrics.…”

Section: Related Workmentioning

confidence: 99%

“…Different segmentation algorithms are implemented in Arabic handwriting recognition [5][6][7][8]. Segmentation problems include over-segmentation, under-segmentation, misplacedsegmentation, broken and touching characters, overlapping, and ligatures.…”

Section: Arabic Handwriting Word-segmentation Algorithmmentioning

confidence: 99%

“…This particular case of the WC is shown in Figure 11 [8]. If the WC precedes the SP, and this WC has an EP under the baseline, then it is rejected.…”

Section: Paw Segmentationmentioning

confidence: 99%

See 2 more Smart Citations

Enhanced technique for Arabic handwriting recognition using deep belief network and a morphological algorithm for solving ligature segmentation

2018

View full text Add to dashboard Cite

Arabic handwriting segmentation and recognition is an area of research that has not yet been fully understood. Dealing with Arabic ligature segmentation, where the Arabic characters are connected and unconstrained naturally, is one of the fundamental problems when dealing with the Arabic script. Arabic character‐recognition techniques consider ligatures as new classes in addition to the classes of the Arabic characters. This paper introduces an enhanced technique for Arabic handwriting recognition using the deep belief network (DBN) and a new morphological algorithm for ligature segmentation. There are two main stages for the implementation of this technique. The first stage involves an enhanced technique of the Sari segmentation algorithm, where a new ligature segmentation algorithm is developed. The second stage involves the Arabic character recognition using DBNs and support vector machines (SVMs). The two stages are tested on the IFN/ENIT and HACDB databases, and the results obtained proved the effectiveness of the proposed algorithm compared with other existing systems.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Arabic Handwriting Word-segmentation Algorithmmentioning

confidence: 99%

See 1 more Smart Citation

Enhanced technique for Arabic handwriting recognition using deep belief network and a morphological algorithm for solving ligature segmentation

2018

View full text Add to dashboard Cite

show abstract

“…The accuracy of segmentation improved from 82.11% to 93.16%. Lawgali etc proposed an algorithm which was relied heavily on the horizontal and vertical projection method in breaking up words into sub-words and characters [50]. During the research, a lot of overlapping characters were lost and considered as noise so that were removed from the final result.…”

Section: Figure 4 Handwritten Arabicmentioning

confidence: 99%

“…The presented method has shown considerable improvement over the projection profile method which was commonly used to segment sub-words or PAWs. images(about 160 document pages) ， set of 20 pages from the dataset is used in the experiments [50] Proposed an algorithm which was relied heavily on the horizontal and vertical projection method in breaking up words into sub-words and characters tested using 800 handwritten Arabic words taken from the IFN/ENIT database NG [51] Word extraction is based on an adaptation of gap metrics and clustering algorithm to identify segmentation thresholds as "within word" or "between words" gaps NG 84.8% correct word extraction…”

Section: Figure 4 Handwritten Arabicmentioning

confidence: 99%