Segmenting Arabic Handwritten Documents into Text lines and Words

International Journal of Advanced Computer Science and Applications

Character segmentation in Unconstrained Arabic handwriting is a complex and challenging task due to the overlapping and touching of words or letters. Such issues have not been widely investigated in the literature. Addressing these issues in the segmentation stage reduces errors in the segmentation process, which plays a significant role in enhancing the accuracy of the Arabic optical character recognition. Therefore, this paper proposes a hybrid approach to improve the accuracy for interconnection, overlapping or touching character segmentation. The proposed method includes several stages: removing extra shapes such as signatures from the document. Using morphological operations, connected components and bounding box detection, detect and extract individual words directly from the document. Finally, the touching characters segmentation is achieved based on background thinning and computational analysis of the word's region. The proposed method has been tested on KHATT, IFN/ENIT database and our own collected dataset. The experimental results showed that the proposed method obtained high performance and improved the accuracy compared to other methods.

Automatic off-line Arabic handwriting recognition still faces a big challenges. Due to the cursive nature of the Arabic language, most of published works are based on recognition of a whole word without segmentation. This paper presents a new framework for the recognition of handwritten Arabic words based on segmentation. This framework involves two phases (training phase and testing phase). In the training phase, Arabic handwritten characters were trained to be recognized, while in the testing phase, words were segmented into characters for recognition. Classification is achieved in two steps (classification of the segmented characters and classification of the word). A dictionary is constructed and used to correct any errors occurring during the previous stages of the recognition process. This work has been tested with IFN/ENIT database and a comparison made against some existing methods and promising results have been obtained.

2016

The greatest difficulty facing the recognition of Arabic handwritten words is segmentation, because Arabic handwriting is cursive with complex multi-form styles. Hence, intensive research efforts are needed to reach an effective Arabic handwriting segmentation system. This paper presents a system which uses morphological features of the Arabic characters for segmentation. The proposed system segments non-overlapped (horizontally connected -e.g. "حسن") as well as overlapped (vertically connected - e.g. "نجد") characters. The result is not very good one. However, it arrives at good directives for more research. As the writing was freely without any restrictions, both over-segmentation and under-segmentation problems affect the system.

In the literature, two methods for the extraction zones of the document are more used. The first method is based on the Mathematical Morphology (MM). The second is based on Hough Transform (HT). The main contribution of this paper is the application of these methods to extract the handwritten components of the complex document. The second contribution is the combination between the HT and the MM. The third contribution is the use of these three developed methods to automatically extract the handwritten components from CENPARMI bank check: numerical amount, literal amount and date zone. We present a concept for automatic evaluation of the results, based on a label tools for the different part of the used documents. We achieve a correct extraction rate of 98.27% for numerical amount, 91.82% for literal amount, and 99.63% for date, extracted by hybrid method HT-MM.

Mobile Multimedia/Image Processing, Security, and Applications 2012, 2012

This paper is concerned with pre-processing and segmentation tasks that influence the performance of Optical Character Recognition (OCR) systems and handwritten/printed text recognition. In Arabic, these tasks are adversely effected by the fact that many words are made up of sub-words, with many sub-words there associated one or more diacritics that are not connected to the sub-word's body; there could be multiple instances of sub-words overlap. To overcome these problems we investigate and develop segmentation techniques that first segment a document into sub-words, link the diacritics with their sub-words, and removes possible overlapping between words and sub-words. We shall also investigate two approaches for pre-processing tasks to estimate sub-words baseline, and to determine parameters that yield appropriate slope correction, slant removal. We shall investigate the use of linear regression on sub-words pixels to determine their central x and y coordinates, as well as their high density part. We also develop a new incremental rotation procedure to be performed on sub-words that determines the best rotation angle needed to realign baselines. We shall demonstrate the benefits of these proposals by conducting extensive experiments on publicly available databases and in-house created databases. These algorithms help improve character segmentation accuracy by transforming handwritten Arabic text into a form that could benefit from analysis of printed text.

The accuracy of handwritten word segmentation is essential for the recognition results; however, it is extremely complex task. In this work, an enhanced technique for Arabic handwriting segmentation is proposed. This technique is based on a recent technique which is dubbed in this work the base technique. It has two main stages: over-segmentation and neural-validation. Although the base technique gives promising results, it still suffers from many drawback such as the missed and bad segmentation-points(SPs). To alleviate these problems, two enhancements has been integrated in the first stage: word to sub-word segmentation and the thinned word restoration. Additionally, in the neural-validation stage an enhanced area concatenation technique is utilized to handle the segmentation of complex characters such as .س Both techniques were evaluated using the IFN/ENIT database. The results show that the bad and missed SPs have been significantly reduced and the overall performance of the system is increased.

SN Computer Science, 2020

The issue of handwritten recognition in Arabic script nature has attracted many researchers from both academic and industrial fields. But their efforts have not reached satisfying outcomes till now. In this paper, a survey has been done in segmentation and recognition of handwritten documents in Arabic script. Most of the previous published works have been analyzed, and some remedies have been suggested. Various strategies used for creating a powerful recognition system have been summarized. This paper presents various algorithms with respect to text, word and characters segmentation and recognition of Arabic document. It analyzes the recognition stage of Arabic script depending on segmentation strategies.

Automatic off-line Arabic handwriting recognition still faces a big challenges. Due to the cursive nature of the Arabic language, most of published works are based on recognition of a whole word without segmentation. This paper presents a new framework for the recognition of handwritten Arabic words based on segmentation. This framework involves two phases (training phase and testing phase). In the training phase, Arabic handwritten characters were trained to be recognized, while in the testing phase, words were segmented into characters for recognition. Classification is achieved in two steps (classification of the segmented characters and classification of the word). A dictionary is constructed and used to correct any errors occurring during the previous stages of the recognition process. This work has been tested with IFN/ENIT database and a comparison made against some existing methods and promising results have been obtained.

2011

A precise and efficient segmentation for handwritten Arabic text is a vital prerequisite for the accuracy of the subsequent recognition phase. In this paper, we present a dualphase segmentation approach. The proposed approach starts first by detecting and resolving sub-words overlapping, then a topological features based segmentation is applied by means of a set of heuristic rules. Because of its crucial importance, the segmentation phase is preceded by a handwritten specific preprocessing phase, that considers issues like word’s skewand slantcorrection. The proposed approach has been successfully tested on a database of handwritten Arabic words, that contains more than 3000 words images. The results were very promising and indicating the efficiency of our approach. KeywordsArabic Handwriting Segmentation, Handwriting Topological Features, Pattern Recognition.

The world heritage of handwritten Arabic documents is huge however only manual indexing and retrieval techniques of the content of these documents are available. To facilitate an automatic retrieval of such hand-written Arabic document, a number of automatic recognition systems for handwritten Arabic words have been proposed. Nevertheless, these systems suffer from low recognition accuracy due to the peculiarities of the handwritten Arabic language. Thus, in this Paper we propose a segmentation based recognition system for handwritten Arabic words. We divide a handwritten word into smaller pieces of a word and then these small pieces are segmented into candidate letters. These candidate letters are converted into their correspondence chain-code representation. Thereafter we extract discrete, statistical and structural features for classifica-tion. Additionally, we introduce a novel active contour based feature to increase the recognition accuracy of strongly deformed Arabic letters....