ABSTRACT

The considerable emergence of new technologies in these last years has led to a large appearance of devices such as digital pen, Tablet Personal Computer and smartphones, which are able to produce online handwriting documents. Thus, the growth of these documents raises the question for identifying the author of a given document. In the last decade, many researches focused on the studies of writer identification task, that can be useful in a variety of applications such as digital rights management in the financial sphere, control access and document analysis. The field of research on online writer identification from Arabic handwriting remains yet an under-exploited compared to Latin scrip. In order to advance the research and development of online Arabic handwritten text and to compare the performance of recent advances on writer identification systems, we organize the ASAR2021 Competition on Online Arabic Writer Identification. A report on the competition will be published in the proceedings of ASAR 2021.The report will comprise description of the participating methods, the evaluation protocol and the final rankings of the participating algorithms. The results of the competition will also be presented in a dedicated session at ASAR 2021.

 

REGISTRATION

Participants need to register to take part in the competition. Please, fill up this form https://forms.gle/7ywxE2EXjfeYcmRi7 including your information.

 

Instructions

The contest aims to promote innovation in online Arabic writer identification, as well as to provide objective and fair comparisons among methods. The ranking of participants is based on writer identification rates with respect to test samples. Participants will be given a limited time to submit their results after the competition started. The contest will consist of two phases:

• Phase 1: Registered participants (https://forms.gle/7ywxE2EXjfeYcmRi7) are provided with training and validation data extracted from ADAB Database to train their algorithms.

• Phase 2: Registered participants receive the test data (without the ground-truth). They must submit their results within two weeks from the release of the test data with a short description of the approach used.

 

Data description

• The training data consists of 3000 online handwritten words document which have been written by 60 different writers. Each writer has contributed 50 handwritten words, while the validation data contains 600 online handwritten words document of the 60 different writers. The data extracted from ADAB Database (An InkML file including trajectory information).  The folder number in the data represents the writer index.

• The test data consists of 170 tests (each test contains 10 online handwritten words document written by one of the 60 writers).

 

The ADAB database (The Arabic handwriting Data Base) was developed to advance the research and development of Arabic on-line handwritten systems. This database is developed in cooperation between the Institut fuer Nachrichtentechnik (IfN) and Research Groups in Intelligent Machines, University of Sfax, Tunisia. The text written is from 937 Tunisian town/village names.

All documents and papers that uses the ADAB database will acknowledge the use of the database by including an appropriate citation to the following:

​

[1] H. Boubaker, A. Elbaati, N. Tagougui, M. Kherallah, H. Elabed, and A. M. Alimi, “Online Arabic Databases and Applications,” Book chapter in : Märgner, V., El Abed, H. (eds) “Guide to OCR for Arabic Scripts,” Chp. Part IV pp 541 – 557 , 2012 , Springer.

[2] M. Kherallah, N. Tagougui, A. M. Alimi, H. El Abed and V. Margner, “Online Arabic handwriting recognition competition”, In International Conference on Document Analysis and Recognition, pp. 1454-1458, 2011.

[3] H. El Abed, M. Kherallah, V. Margner and A. M. Alimi, “on-line Arabic handwriting recognition competition ADAB database and participating systems”, International Journal on Document Analysis and Recognition, volume 13, July 2010.

 

EVALUATION AND RANKING

Participants will be asked to provide results in the form of a xlsx file (https://drive.google.com/file/d/1RSOWgp0BcyTaKCAAunuKyrDIA-XFk8gu/view?usp=sharing).

The first column should be the IDs of the writers (60). The 170 next columns should correspond to the score of the tests respectively. Participants must provide the score of each test to each writer given that each test will concern 10 online handwritten words document written by one of the 60 writers. The writer with the maximum score is the final answer.

An example of the expected file format is shown in the following.

​​