Abstract—In this work we present an off line system for the recognition of the Arabic handwritten words of the Algerian departments. The study is based mainly on the evaluation of neural network performances, trained with the gradient back propagation algorithm. The used parameters to form the input vector of the neural network are extracted on the binary images of the handwritten word by several methods: the parameters of distribution, the moments centered of the different projections and the Barr features. Let's note that these methods are applied on segments gotten after the division of the binary image of the word in six segments. The classification is achieved by a multi layers perceptron. Detailed experiments are carried and satisfactory recognition results are reported. Index Terms—Optical characters recognition, neural networks, barr features, image processing, pattern recognition, features extraction. I. INTRODUCTION Writing, which has been the most natural mode of collecting, storing, and transmitting information through the centuries, now serves not only for communication among humans but also serves for communication of humans and machines. The handwritten writing recognition has been the subject of intensive research for the last three decades. However, the early researches were limited by the memory and power of the computer available at that time. With the explosion of information technology, there has been a dramatic increase of research in this field. The interest devoted to this field is explained by the potential mode of direct communication with computers and the huge benefits that a system, designed in the context of a commercial application, could bring. According to the way handwriting data is generated, two different approaches can be distinguished: on-line and off -line. In the former, the data are captured during the writing process by a special pen on an electronic surface. In the latter, the data are acquired by a scanner after the writing process is over. Off-line and on-line recognition systems are also discriminated by the applications they are devoted to. The off-line recognition is Manuscript received September 2, 2013; revised October 20, 2013. This work was supported in part by the Skikda Electronic Laboratory, skikda university, Algeria and the Electrical Engineering Department El Jouf University, Arabie Saoudite. Salim Ouchtati is with the Skikda Electronic Laboratory, Engineering Department, Faculty of Technology, Skikda University, Route El Hadaik, Bp: 26 Skikda 21000, Algeria (Phone: 0021393935198, e-mail: ouchtatisalim@ yahoo.fr). Mohammed Redjimi is with the Computer Science Department, Faculty of Science, Skikda University, Algeria (e-mail: redjimimed@yahoo.fr). Mouldi Bedda is with the Electrical Engineering Department, College of Engineering Al-Jouf University, Arabie Saoudite (e-mail: mouldi_bedda@yahoo.fr). dedicated to bank check processing, mail sorting, reading of commercial forms, etc., while the on-line recognition is mainly dedicated to pen computing industry and security domains such as signature verification and author authentication. Optical characters recognition is one of the successful applications of handwriting recognition; this field has been a topic of intensive research for many years. First only the recognition of isolated handwritten characters was investigated [1], [2], but later whole words were addressed [3]. Most of the systems reported in the literature until today consider constrained recognition problems based on vocabularies from specific domains, e.g. the recognition of handwritten check amounts [4], [5] or postal addresses [6], [7]. Free handwriting recognition, without domain specific constraints and large vocabularies, was addressed only recently in a few papers. The recognition rate of such systems is still low, and there is a need to improve it. Character and handwriting recognition has a great potential in data and word processing, for instance, automated postal address and ZIP code reading, data acquisition in banks, text-voice conversions, etc. As a result of intensive research and development efforts, systems are available for English language [8]-[10], Chinese language [11], Arabic language [12], [13] and handwritten numerals [14]. There is still a significant performance gap between the human and the machine in recognizing unconstrained handwriting. This is a difficult research problem caused by huge variation in writing styles and the overlapping and the intersection of neighboring characters. Today, the OCR (Optical Characters Recognition) systems are only able to recognition high quality printed or neatly handwritten documents. The current research is now basing on documents that are not well handled and including severely degraded, omnifont machine printed text, and unconstrained handwritten text. A wide variety of techniques are used to perform handwriting recognition. II. THE DIFFERENT PARTS OF THE REGONITION SYSTEM In the setting of the handwritten writing recognition, we proposed an off line system for the recognition of the handwritten Arabic words of the Algerian departments (shown in the Fig. 1), this system is divided in three phases:  Acquisition and preprocessing  Feature extraction  Recognition A. Acquisition and Preprocessing 1) Acquisition Before analyzing the different processing steps, let's recall Recognition of the Arabic Handwritten Words of the Algerian Departments Salim Ouchtati, Mohammed Redjimi, and Mouldi Bedda International Journal of Computer Theory and Engineering, Vol. 6, No. 2, April 2014 129 DOI: 10.7763/IJCTE.2014.V6.850