Afef Kacem

A novel approach for the recognition of a wide Arabic handwritten word lexicon

This paper introduces a novel approach for the recognition of a wide vocabulary of Arabic handwritten words. Note that there is an essential difference between the global and analytic approaches in pattern recognition. While the global approach is limited to reduced vocabulary, the analytic approach succeeds to recognize a wide vocabulary but meets the problems of word segmentation especially for Arabic. Combining the neural approach with some linguistic characteristics of the Arabic, it is expected that we become able to recognize better and to handle a large vocabulary of Arabic handwritten words. The proposed approach invokes two transparent neural networks, TNN_1 and TNN_2, to respectively recognize roots, schemes and the elements of conjugation from the structural primitives of the words. The approach was evaluated using examples from a database established for this purpose. The results are promising, and suggestions for improvements are proposed.

EXTRAFOR: automatic EXTRAction of mathematical FORmulas

We present a method for automatic extraction of mathematical formulas from images of documents without character recognition. Formula extraction is first done by location of its most significant symbols, then extension to adjoining symbols using contextual rules until delimitation of the whole formula space. Mathematical symbol labelling is realised from models created at the learning stage using fuzzy logic. From the experiments, we found that the average rate of primary labelling of mathematical symbols is about 95.3%. The obtained results have demonstrated the applicability of our system since 90% of mathematical formulas are well extracted from documents printed with high quality.

Towards Arabic Handwritten Word Recognition via Probabilistic Graphical Models

In this work, we propose a novel system for the recognition of handwritten Arabic words. It is evolved based on horizontal-vertical Hidden Markov Model and Dynamic Bayesian Network Model. Our strategy consists of looking for various HMM architectures and selecting those which provide the best recognition performance. Experiments on handwritten Arabic words from IFN/ENIT strongly support the feasibility of the proposed approach. The recognition rates achieve 92.19% with horizontal-vertical Hidden Markov Model and 88.82% with a Dynamic Bayesian Network.

Structural Features Extraction for Handwritten Arabic Personal Names Recognition

Due to the nature of handwriting with high degree of variability and imprecision, obtaining features that represent words is a difficult task. In this research, a features extraction method for handwritten Arabic word recognition is investigated. Its major goal is to maximize the recognition rate with the least amount of elements. This method incorporates many characteristics of handwritten characters based on structural information (loops, stems, legs, diacritics). Experiments are performed on Arabic personal names extracted from registers of the national Tunisian archive and on some Tunisian city names of IFN-ENIT database. The obtained results presented are encouraging and open other perspectives in the domain of the features and classifiers selection of Arabic Handwritten word recognition.

A System for an Automatic Reading of Student Information Sheets

In this paper we present a student information sheet reading system. Relevant algorithm is proposed to locate and label handwritten answer field. As information sheets can be filled in Arabic and/or in French, automating the script language differentiation is a pre-recognition required in the proposed system. We have developed a robust and fast field classification and script language identification method, based on a decision tree, to make these processing practical for sheet recognition. To this end, the system uses several novel features (loops, descenders, diacritics) and analyses the lower profile of script. The classification rates are 92.5% for numeric fields, 94.34% for Arabic scripts and 94.66% for French scripts. Experimental results, carried on 80 sheets, show our system provides an effective way to convert printed sheets into computerized format or collect information for database from printed sheets.

Embedded formulas extraction

A new approach for separating mathematics from usual text is presented. Contrary to the existing methods, it is more oriented toward the segmentation than the recognition, isolating the formulas outside and inside the text lines. The objective is to delimit a part of text which could disturb the OCR application, not yet trained for formula recognition and restructuring. The method is based on an adaptive segmentation working at two levels (1) A primary labelling identifies the more characteristic symbols; (2) A secondary labelling extends the context of the symbols for delimiting the formula inside the text. Experiments done on some commonly seen mathematical documents, show that our proposed method can achieve quite satisfactory rate making mathematical formulas extraction more feasible for real-world applications. The average rate of primary labelling of mathematical symbols is about 95.3% and their secondary labelling can improve the rate about 4%. Thus, about 95% of formulas are well extracted from images of documents printed with high quality.

Co-occurrence Matrix of Oriented Gradients for word script and nature identification

In this paper, we propose a new scheme for script and nature identification. The objective is to discriminate between machine-printed/handwritten and Latin/Arabic scripts at word level. It is relatively a complex task due to possible use of multi-fonts and sizes, complexity and variation in handwriting. In the proposed script identification system, we extract features from word images using Co-occurrence Matrix of Oriented Gradients (Co-MOG). The classification is done using different classifiers. Extensive experimentation has been carried on 24000 words, extracted from standard databases. An average identification accuracy of 99.85% is achieved by k Nearest Neighbors (k-NN) classifier which clearly outperforms results of some existing systems.

Segmentation of Touching Component in Arabic Manuscripts

Touching components are connection zones occurring between text-lines or words of the same line and are one of the problems that make unconstrained handwritten text segmentation greatly hard. In this paper, we propose a recognition based method to separate these components once localized in Arabic manuscript images. It first identifies, for a given touching component, a similar model stored in a dictionary with its correct segmentation, using shape context descriptor and an interpolation function. Then, it segment the touching component based on the distance from the midpoints of the identified models parts. Tests are performed using a database of touching components and two metrics: Manhattan and Euclidean distances. Experimental results show the effectiveness of the proposed segmentation method.

A proposal for touching component segmentation in Arabic manuscripts

Text-line segmentation is one of the key factors which affect the performance of handwriting recognition system. Therefore, to make recognition systems more effective and accurate, segmentation of touching text-lines is an important task. One of the problems making this task crucial is the presence of touching components (TCs) representing connections between word letters of consecutive text-lines or those of words of the same text-line. The proposed method aims to segment TCs. It is mainly based on two steps: (1) finding for a localized TC a similar model, stored in a dictionary with its correct segmentation, using shape context descriptor and an interpolation function: the thin plate spline transformation, (2) segmenting the TC based on central point of the found similar model parts. TCs are assumed to be already extracted from Arabic manuscript images. Experiments are carried on a common TC database, using two metrics: Manhattan and Euclidean distances. Obtained results outperform the state of the art, considering the different types, variability and complexity of the TCs data set, and show the effectiveness of the proposed TC segmentation method.

A proposal system for historic Arabic manuscript transcription and retrieval.

In this paper, we propose a computer-assisted transcription system of old registers, handwritten in Arabic from the 19th century onwards, held in the National Archives of Tunisia (NAT). The proposed system assists the human supervisor to complete the transcription task as efficiently as possible. This assistance is given at all different recognition levels. Our system addresses different approaches for transcription of document images. It also implements an alignment method to find mappings between word images of a handwritten document and their respective words in its given transcription.