ICDAR'2013 Competition on On-Line Arabic Handwriting Recognition


Monji Kherallah, Najiba Tagougui and Adel M. AlimiEmails:{monji.kherallah,najiba.tagougui, adel.alimi }@ieee.org


REGIM: Research Group on Intelligent Machines.

University of  SfaxNational  School of Engineers (ENIS) BP "W", Sfax, 3038,  Tunisia


Haikal El Abed and Volker MÃrgnerEmails:{ el-abed , maergner }@ifn.ing.tu-bs.de

Institute for CommunicationsTechnology (IfN),  Braunschweig Technical UniversityGermany


Volker Märgner: received his diploma (Dipl.-Ing.) and doctorate (Dr.-Ing.) degrees in electrical engineering from the Technische Universität Carolo Wilhelmina zu Braunschweig (TUBS), Germany, in 1974 and 1983 respectively. Since 1983, he has been working at the TUBS. Currently he is a member of the research and teaching staff at the Institute for Communications Technology, in the position of an academic director. He lectures in image processing and pattern recognition. His main areas of research are image processing and pattern recognition. Currently, he is working on image pre-processing and pattern recognition methods and their application to industrial quality control as well as to the recognition of cursive handwriting on documents. He developed recognizer for printed German text and for German handwritten words. Robust pre-processing and feature extraction with an HMM based recognizer are the key features of this solution. Since 1991 he is also working on Arabic text recognition, at the beginning on printed text recognition thereafter on handwritten Arabic word recognition. This work is done in close cooperation with Tunisian universities. He developed the IFN/ENIT-database of handwritten words in 2002 and organizes the biennial competition on Arabic handwriting recognition within the ICDAR conference since 2005. He worked on the important task of system evaluation, in particular on the evaluation of document segmentation results. He published more than 60 papers including journal papers and book chapters. He is a member of program committees of conferences and workshops. He is a reviewer for international journals, including IEEE-PAMI, IJDAR, and PR and he is a member of VDE/VDI, DAGM, IAPR (TC10, TC11), and IEEE.


Adel M. Alimi: was born in Sfax ( Tunisia) in 1966. He graduated in Electrical Engineering 1990, obtained a PhD and then an HDR both in Electrical & Computer Engineering in 1995 and 2000 respectively. He is now professor in Electrical & Computer Engineering at the University of Sfax. His research interest includes applications of intelligent methods (neural networks, fuzzy logic, evolutionary algorithms) to pattern recognition, robotic systems, vision systems, and industrial processes. He focuses his research on intelligent pattern recognition, learning, analysis and intelligent control of large scale complex systems. He is associate editor and member of the editorial board of many international scientific journals (e.g. "Pattern Recognition Letters", "NeuroComputing", "Neural Processing Letters", "International Journal of Image and Graphics", "Neural Computing and Applications", "International Journal of Robotics and Automation", "International Journal of Systems Science",etc.). He was guest editor of several special issues of international journals (e.g. Fuzzy Sets & Systems, Soft Computing, Journal of Decision Systems, Integrated Computer Aided Engineering, Systems Analysis Modelling and Simulations). He was the general chairman of the International Conference on Machine Intelligence ACIDCA-ICMI'2005 & 2000. He is an IEEE senior member and member of IAPR, INNS and PRS. He is the 2009-2010 IEEE Tunisia Section Treasurer, the 2009-2010 IEEE Computational Intelligence Society Tunisia Chapter Chair, the 2011 IEEE Sfax Subsection, the2010-2011 IEEE Computer Society Tunisia Chair, the 2011 IEEE Systems, Man, and Cybernetics Tunisia Chapter, the SMCS corresponding member of the IEEE Committee on Earth Observation, and the IEEE Counselor of the ENIS Student Branch.


Monji Kherallah: was born in Sfax, Tunisia, in 1963. He received the Ing. Diploma degree and the Ph.D both in electrical engineering in 1989 and 2008 from University of Sfax (ENIS). For fourteen years ago, he was an engineer in Biotechnologie Center of Sfax. Now he teaching in Faculty of Science of Sfax and member in Research Group of Intelligent Machines: REGIM. His research interest includes the Handwritten Documents Analysis and Recognition. The techniques used are based on intelligent methods, such as neural network, logic fuzzy, genetic algorithm etc. He is one of the developers of the ADAB-Database (used by more than 50 research groups from more than 10 countries). He co-organized the Arabic Handwriting Recognition Competitions at the Online Arabic Handwriting Competitions at ICDAR 2009 and ICDAR 2011. He has more than 40 papers, including journal papers and book chapters. He is a member of IEEE and IEEE AESS Tunisia Chapter Chair, 2010 and 2011. He is reviewer of several international journals.


Haikal El Abed: is a Senior Research Engineer at the Braunschweig Technical University, Germany.

Since 2001, he has been working at the Institute for Communications Technology (IfN), Department of Signal Processing for Mobile Information Systems. He has specialized in image and signal processing, document analysis systems design and configuration, and Arabic/Latin manuscripts recognition. He coordinated different national and international research projects and is one of the developers of the IfN/ENIT-Database (used by more than 100 research groups from more than 30 countries). He co-organized the Arabic Handwriting Recognition Competitions at the ICDAR 2005, 2007, 2009, and 2011, and the Handwriting Competition and the Online Arabic Handwriting Competitions at ICDAR 2009 and ICDAR 2011. He is competition chair at the ICFHR 2010 in India, ICDAR 2011 in China and ICFHR 2012 in Italia. He has organized different special sessions with the topics Arabic\Latin handwriting\printed recognition technologies on international conferences and was invited for different talks. He has more than 50 papers, including journal papers and book chapters. He is a member of IEEE, DAGM, IAPR (TC-10 and TC-11), and VDE/VDI and a frequent reviewer for international journals, including the IEEE Transactions on Pattern Analysis and Machine Intelligence and Pattern Recognition Letters. He is a member of the program committee of different international conferences and workshops.

Najiba TAGOUGUI graduated in Computer Sciences in 2005, obtained a master degree in New technologies of dedicated computer systems in 2007. She is now a PhD student in Computer Systems Engineering at the University of Sfax. His research interest includes applications of intelligent methods to pattern recognition. She focuses her research on intelligent pattern recognition especially Arabic Handwriting Recognition. She is an IEEE student member.


Our proposal: On line recognition of the cursive Arabic handwritten words, aims to contribute in the evolution of on line Arabic handwriting recognition research. Since 2009, the freely available ADAB data base is used by some groups all over the world to develop on line Arabic handwriting recognition systems. This database was adopted successively in ICDAR 2009 and ICDAR 2011 competitions for learning and testing the systems competing in the recognition of the online cursive Arabic handwritten words. Respect to these last competitions, the proposed edition for the ICDAR 2013 online handwriting competition is characterised by the extension of the ADAB database (version 3.0) by a new dataset (set 7) collected on white board by young elementary student. The objective is to favourite the recognition systems that are less sensitive to the handwriting trajectory acquisition tool. Thus the test dataset (set 8) is composed by 3500 samples collected by different digital ink device : tablet PC, white board and smart phone.

Evaluation Process

The object is to run each Arabic handwritten word recognizer (trained on a part of version 3.0 of the ADAB-database) on an already published part of the ADAB-database and on a test set not included in the published part. The recognition results on word level of each system are compared on the basis of correct recognized words, i.e. there correspondent consecutive Numeric Character References (NCR). A dictionary can be used in the recognition process. A recognizer may return up to 10 candidates for each classification that not only the first ranked result can be used for comparison but also the correct result between the 5 or 10 candidates will be used for comparison.

Running a Recognizer

Note that the evaluation process of all systems will be released in our laboratory REGIM: Group of Research on Intelligent Machines. We run your recognizer (called myrec) by invoking it from the command line as follows:myrec input.txt output.txt


The input file is just a list of relative paths to each *.inkml online trace to be recognized. For example:




Word/ 1232017605345.inkml

Word/ 1232017618668.inkml




The output file should have one line as result for each input file. Each line should show the name of the online trace file that was recognized, followed by the responses (sequence of NCRs code) for that file.

Each response is given as a pair of values: the text, followed by the confidence. In the following example the first line shows that for the file Word/ 1232017605345.inkml the recognizer has produced three word hypotheses with confidences of 1.0,0.5 and 0.2 respectively.

Word/ 1232017605345.inkml U0642 U0631 U0639 U0629 U0020 U0627 U0644 U0646 U0651 U0627 U0638 U0648 U0631 1.0 U0633 U0648 U0633 U0629 U0020 U0627 U0644 U0632 U0651 U0647 U0648 U0631 0.5 U0637 U0628 U0631 U0642 U0629 U0020 U0627 U0644 U0645 U0637 U0627 U0631 0.2


Word/ 1232017618668.inkml U0627 U0644 U0645 U0647 U062F U064A U0629 U0020 U0627 U0644 U0632 U0651 U0647 U0631 U0627 U0621 0.7 U0627 U0644 U0631 U0642 U0651 U0648 U0628 U0629 U0020 U0627 U0644 U062D U0645 U0631 U0627 U0621 0.3


Important dates

Deadline for submission of systems:   Marsh 15th, 2013