To input handwriting characters

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Chapter 1


1.1 Background

1.1.1 Tablets

A slate like shape device uses for mobile computers which contain a pen like devices used to input handwriting characters or text directly in to the computer to operate the computer. By using tablets is more mobile way to communicate the PC or other computing devices. The main focus of using tablets pc is mobility. The drivers or software for tablets Pc are targeted particularly for Microsoft tablets API, if we want to run other operating system we need a software of tablets for this environments

1.1.2 MATLAB7

MATLAB is a powerful computational tool for high performance language and other technical computing, this is user friendly environment which brings together programming visualization and other computing process and it use mathematical notation for problems and their solutions. The development of computation algorithm, maths and data Acquisition, engineering and scientific visualization, modeling and simulation data analysis and prototyping and other application development including GUI and all data and information are store in the form of array and to need for dimensioning. Which is used to solve a lot technical computing problems.

The name MATLAB stands for matrix laboratory. It was basically for matrix computation but nowadays it is used by students in universities for the mathematical, scientific, engineering tool. In industry it is used for research, analysis and other development process.

MATLAB contains a Varity of toolboxes for specific solution with specific technology .These toolboxes used to learn and implements specific technology, Toolboxes contain Matlab function or M files which are used to solve particular classes and their problems. The fields in which these M files or toolboxes includes Fuzzy logic, simulation image processing, signal processing, control system, natural networks and many more.

MATLAB comprises the following parts.

Matlab Language:

Matlab language is a high level array language which contains input-output, object oriented feature, control and selection statements, data structure etc ,it used to create a small programs or large complex application,

The MATLAB Mathematical Function Library.

Matlab is contains the vast collection of algorithms and other mathematical function like average, sum, sine etc and complex function such as matrix eigenvalues. Fourier transforms etc.

Development Environment.

This is use as integrated development environment (IDE) for many languages and provide tools which facilities the user to use Matlab function or m files." It includes the MATLAB desktop and Command Window, a command history, an editor and debugger, and browsers for viewing help, the workspace, files, and the search path"


Matlab contain M files for 2 D and 3D data visualization, graphical representation, images processing and animation, it contain high-level function for videos processing and other graphical operation on images or videos. By Matlab we can use create our graphical user interface and also customize the appearance of graphics.

The MATLAB Application Program Interface (API).

Matlab is use to code a FORTRAN and a C program that interact with Matlab. It provides a library or MAT-files which is used for dynamic linking of Matlab with other language.

Matlab documentation; it facilitate to documents in online or printed format and provide to learn all its features. all the codes used the proposed work is created and compiled in Matlab. Image processing toolbox provides some build in function M-files which are helpful in this preprocessing and Feature Extraction. I can also use my own created functions and M-files in the feature Extraction of character images.

1.1.3 Arabic Character

Arabic language is an ancient language and spoken by millions of people in world. Arabic words and text are used thousands of year ago, it is used and understood by people in world, it is the nation and official languages of many Asian and African countries. with the advancement of technology works are started to adapt Arabic script to easy and understandable use in computers with out using the integrity if Arabic alphabets. one of this task is recognizing the Arabic characters and text, this efforts suffer many usual problems due to different features and characteristic of Arabic text.

The Arabic letters are written from left to right and contain 28 letters. According to position and shape of the letters like beginning ,middle and ending or isolated as well as according to the way of writing(writing style) NASTALEEQ, NASEKH etc the Arabic letters can take two more different shapes, and some of the letters comprises secondary object like dots this makes Arabic characters more complicated to recognize, the other reason that Arabic characters difficult to recognize is the different shape of same letter and the similarities of different characters like the characters Baa,Taa,Saa are three letters in Arabic language having similar body, but their difference is the position of dots objects which lies below are above the letters. The characters like jeem, Hhaa, Khaa,Aain and Ghain are Arabic characters, the difference among is the is position of the dots objects. Some character is Arabic are completely different from other like Haa, in Arabic language the character Haa has completely 3 different shape in different positions as shown in the given figures. The shapes, dots objects and positing makes the Arabic text difficult to process and recognize.

Arabic characters contain at least one stroke. The strokes will be the starting point or pixels turning pixel and ending pixels. The starting pixels is primary strokes and others is called secondary strokes, using these strokes and their position causes the recognition of characters

1.1.4 Handwritten Arabic Characters Recognition (HACR)

The HACR is a technique used to recognize the Arabic characters by comparing the gained features of characters to the original characters. HACR process is done in the following steps.


In this process the characters are retrieves, and the extra information about the characters which are not used in the Recognition stages are removed. In other words their removal are could affect the identity of the characters.

Feature Extraction:

In this stage the characters image hold with the extracted features having useful information about the characters like strokes information, starting and ending points etc. using these features the cursive handwritten Arabic text will be possible to recognize.


The evaluation of the unknown characters extracted features and comparing them with a set feature of possible characters is known as Recognition. The character and the unknown characters are reported similar. we compare the features and generates the recognition rate ,which is used to find or analysis the accuracy and effectiveness. we can then analysis more low level features to improve the recognition.

The finishing step involves storing the output in one of the industry standard formats such as RTF, PDF, WORD and plain UNICODE text Online Character Recognition

Online Character Recognition it is very useful technique for offering a easy assess to human and computer interface it is an automatic conversion letters as it is written on pen-based input devices which contains sensor which sense the pin tip movements like pen up and down movements. This input data is a dynamic representation of handwriting. and the retrieved characters is used in different character recognition processes like pre-processing segmentation feature extraction, recognition etc. Elements involving in the online characters recognition include

  • A tablet PC: The writing environment
  • A pen like device for writing over the tablets
  • An application which sense the movements of the pen and recognize the strokes, and make the character ready for next steps of of character recognition like preprocessing feature extraction etc Offline character recognition

Offline handwritten character recognition is the phase of handwriting character recognition in which the save images with different format which are useable in many applications and computers. The images in types are a static because it is a saved data or present in the storage media. The offline images recognition technology has a wide application in the fields of business like handwritten documents etc in health etc. offline handwritten recognition is difficult because of different ways of writing by different styles. So different techniques and algorithms are used in the recognition of handwritten characters images

1.2 Scope of This Work:

The Project is designed to classify and identify a online data containing Arabic characters using two pace approaches. In the first phase the data as a Arabic character was retrieved from tablets and used some preprocessing steps over the characters like smoothing, tinning etc using Matlab, and in the other phase is feature extraction and recognition. In phase we can get some of the feature of the character like strokes identification starting point ending point's etc, and in recognition by comparing this features with original images using some technique we can recognize the characters.

1.2 Objectives and Applications of This Work:

Handwritten Arabic character Recognition (HACR) is a useful way of mobile communication with computing devices like computers or laptops etc. HACR expends the knowledge to a new approach; the cursive script of Arabic and other language like Urdu Farsi of become available in computing.

The most important objective of HACR is to make computing device to understand human writing or human activities. in character recognition are used in the automation process due to which man and computer interaction are improved.

The interaction of man and machine are in lot of application like banking sector

The ultimate goal of character recognition is to conjure up the human reading capabilities. It is used in data entry operation like e-book shops, data warehouses. Document identification, archives libraries shipment receipt processing and many other uses

Chapter 2

Literature Review

Handwritten Character Recognition:

Since the beginning of writing as a form of communication, paper prevailed as the medium for writing. Electronic media is replacing paper with time. Because it preserves space and is fast to access, electronic media are constantly gaining esteem. The convenience of paper, its pervasive used for communication and archiving, and the quantity of information already on paper, press for quick and accurate methods to automatically read that information and adapt it into electronic form [Albadr95].

The latent application areas of automatic reading machines are numerous. One of the earliest, and most thriving, applications is sorting checks in banks, as the volume of checks that circulates daily has proven to be too huge for manual entry. Other applications are detailed in the next section [Govindan90, Mantas86].

The machine imitation of human reading (i.e. handwritten character recognition) has been the subject of widespread research for more than five decades. Character identification is pattern recognition application with a crucial aim of simulating the human reading capabilities of both machine printed and handwritten cursive text. The currently available systems may interpret faster than humans, but cannot reliably read such a wide diversity of text nor consider context. One can say that a great quantity of further effort is required to, at least, narrow gap between the machines learning and human understanding capabilities. The practical significance of HCR applications, as well as the interesting nature of the HCR problem, has lead to great research interest and assessable advances in this field. Now, commercial HCR systems for Latin characters are commonly accessible on personal computers achieving recognition rates above 99% [McClelland91, Welch93]. Further, systems on the market can now interpret a variety of writing styles (e.g., hand-written, printed Omni-font), and character sets including Chinese, Japanese, Korean, Cyrillic, and Arabic.

Since the 50s, researchers have carried out far-reaching work and published many papers on character recognition. Nearly all of the published work on HCR has been on Latin, Japanese or Chinese characters. This has started since the median 40s for Latin, the middle of the 1960s for Chinese and Japanese. The following are positive surveys and reviews on Latin character recognition. Reference may be made to [Mori92] for historical appraisal of OCR research and development. The survey of [Govindan90] includes surveys of other languages; [Mantas86] has an overview of character identification methodologies, [Impedovo91] on commercial OCR systems, [Tian91] on machine-printed OCR, [Tappert90, Wakahara92] for on-line handwriting identification. [Suen80] has a survey on automatic identification of hand printed characters (viz. numerals, alphanumeric, FORTRAN, and Katakana), while [Nouboud90] produced a review of the recognition of hand-printed (non-cursive) characters and conducted beta tests on a business system. [Bozinovic89, Simon92] surveyed off-line cursive word recognition, Jain et al [Jain2000] reviewed statistical pattern recognition methods, and [Plamondon2000] comprehensive survey of online and offline handwriting identification. Two bibliographies of the fields of HCR and document scrutiny appeared in [Jenkins93, Kasturi92]. [Stallings76, Mori84], produced surveys on identification of Chinese machine- and hand-printed characters, respectively, and Liu et al [Liu2004] addressed the state of the art of online identification of Chinese characters.

Arabic Character Recognition:

Although almost one billion people world-wide, in several diverse languages, use Arabic characters for writing (Arabic, Persian, and Urdu are the most noted examples), Arabic character identification has not been researched as thoroughly as Latin, Japanese, or Chinese. The first published work on Arabic character acknowledgment may be traced back to 1975 by Nazif [Nazif75] in his master's thesis. In his thesis a system for the identification of printed Arabic characters was developed based on extracting strokes that he called radicals (20 radicals are used) and their positions. He used correlation between the templates of the deep-seated and the character image. A segmentation phase was included to segment the cursive text. Years later Badi and Shimura [Badi78, Badi80] and Noah [Nouh80] toiled on printed Arabic characters and Amin [Amin80] on hand-written Arabic characters. Surveys on AOTR may be referred in [Amin85a, Amin98, Shoukry89, Jambi91, Albadr95, Nabawi2000,Ahmad94]

On-line systems are restricted to recognizing hand-written text. Some systems recognize remote characters [Ali89, Amin80, Amin85b, Amin87, ElSheikh89, ElSheikh90b, ElWakil87, ElWakil89, Saadallah85] and hand-written mathematical formulas [ElSheikh90c, Amin91b], while others recognize cursive words [Badi78, Badi80, Badi82, Amin82a, Amin82b, Shaheen90, AlEmami90]. Since the segmentation problem in Arabic is non-trivial the concluding systems deal with much harder problems

While several off-line systems use video cameras to digitize pages of text (e.g., [Abbas86, Goraine92, Amin86, HajHassan85, HajHassan90, Nouh80, Nouh87, Nouh89, Sarfraz2003, Sarfraz2004]), the inclination now is to use scanners with resolutions ranging from 200 to 400 dots per- inch (e.g., [AbdelAzim89c, AbdelAzim90a, AlYousefi88, Amin91a, Bouhlila89, ElDabi90, ElSheikh88a, Ramsis88, Sarfraz2003a, Sarfraz2003b, Zidouri2002, Zidouri2005]). Scanners set up less noise to an image, are less pricey, and more convenient to use for character recognition, especially when coupled with automatic document feeders, automatic Binarization, and image enhancement.

Among the off-line systems that identify hand-written isolated characters are [Abuhaiba90, AlYousefi90, AlTikriti85, ElDesouky92, Hyder88]. [Abbas86, AbdelAzim89b, Goneid92] identify hand-written Arabic (Hindi) numerals, and [Badi80, Badi82, Goraine92, Jambi92, Zahour91] distinguish hand-written words. The majority of off-line systems distinguish typewritten cursive words [AbdelAzim89c, AbdelAzim90a, Bouhlila89, ElDabi90, Amin86, ElKhaly90, ElSheikh88b, Goraine89, Khella92, Margner92, Nazif75, Nouh87, Ramsis88, Tolba89, Tolba90, ElRamly89c, HajHassan90, HajHassan91], while [ElShiekh88a, Mahdi89, Mahmoud94, Nouh80, Nouh89, NurulUla88, Fayek92, Sarfraz2005d, Zidouri2005] identify only typewritten isolated characters. The systems of [Abdelazim90b, AlBadr92, ElGowely90, Kurdy92, Fakir93] are intended to recognize typeset words. One of the systems [Abdelazim89a] recognizes bilingual (Arabic/Latin) typewritten words. Examples of systems for detection of other languages that use Arabic script are [Parhami81, Yalabik88, Hyder88], which are designed for the identification of Persian, Ottoman (Old Turkish), and Urdu, respectively.

Uses of Handwritten Character Recognition:

Optical character recognition technology has many practical applications that are independent of the treated language. The following are some of these applications:


For cataloging bank checks since the number of checks per day has been far too large for Manual arrangement.

Data Processing:

For inflowing data into commercial data processing files, for example inflowing the names and addresses of mail order customers into a database. In addition, it can be worn as a work sheet reader for payroll accounting.

In Postal Department:

For postal address reading, cataloging and as a reader for handwritten and printed postal codes.


Premium typescript may be read by recognition equipment into a computer typesetting system to keep away from typing errors that would be introduced by keypunching the text on computer peripheral equipment.

Use by Blind:

It is used as a reading abet using photo sensor and tactile simulators, and as a sensory aid with sound output. Additionally, it can be worn for reading text sheets and reproduction of Braille originals.

In Facsimile Transmission:

This procedure involves transmission of pictorial data over communications channels. In practice, the pictorial data is mainly text. Instead of transmitting characters in their pictorial representation, a character identification system could be used to recognize each character then transmit its text code. Finally, it is worth to say that the major potential application for automatic character identification is as a general data entry for the automation of the work of an ordinary office typist.

Development of New HCR Techniques:

As HCR and OCR research and development advanced, demands on handwriting identification also increased because a lot of data (such as addresses written on envelopes; sums written on checks; names, addresses, identity numbers, and dollar values written on invoices and forms) were written by hand and they had to be pierced into the computer for processing. But early HCR techniques were based generally on template matching, simple line and geometric features, stroke detection, and the extraction of their derivatives.

Such techniques were not classy enough for practical identification of data handwritten on forms or documents. To cope with this, the Standards Committees in the United States, Canada, Japan, and some countries in Europe designed some handprint models in the 1970s and 1980s for people to write them in boxes [7]. Hence, characters written in such specified shapes did not diverge too much in styles, and they could be recognized more easily by OCR machines, especially when the data were pierced by controlled groups of people, for example, employees of the same company were asked to write their data like the advocated models. Sometimes writers were asked to follow certain bonus instructions to enhance the quality of their samples, for example, write big, close the loops, use simple shapes, do not link characters, and so on. With such constraints, OCR detection of handprint was able to flourish for a number of years.

Recent Trends and Movements:

As the years of exhaustive research and development went by, and with the birth of several new conferences and workshops such as IWFHR (International Workshop on Frontiers in Handwriting Recognition), 1 ICDAR (International Conference on Document Analysis and Recognition), 2 and others [13], identification techniques advanced rapidly. Moreover, computers became much more authoritative than before. People could write the way they normally did, and characters need not have to be written like specified models, and the subject of unimpeded handwriting recognition gained considerable momentum and grew swiftly. As of now, many new algorithms and techniques in pre-processing, feature extraction, and powerful classification methods or technique have been urbanized [8,9]

Chapter 3

Extraction and Recognition of online character

3.1 Stage of online Arabic character Recognition

In online character recognition system the data are as a Arabic character are input through input device called a tablet. After receiving the data certain operation are performed on this data (Arabic character) to extract their feature and recognize these characters by comparing the original character in the database. These extraction and recognition process are done in the following stages

  1. Data Acquisition
  2. Preprocessing
  3. eature extraction
  4. Recognition

3.2 Data acquisition

The retrieval of data from input devices like tablets mouse etc. this received data as a character is used in the character recognition processing. In data acquisition techniques

We can operate certain operation on it

3.2.1 Smoothing

As during the writing in tablets or input device strokes contains zigzag direction due to natural hand movements .this a slow change but it affects recognition process. To overcome this issue data filtering is used data filtering use the simple technique smoothing. During smoothing process the coordinate of the original point use the neighbor pixels when we input data are put by writing in the tablets or other input devices

During writing in the tablets the writers create extra line and other curves in the character. This event happen when pen-down, pen up occurs. Due to which the characters' joints effects and create problem in the detection of joints. So removal of these are necessary which are almost occur at the starting and ending of characters strokes.

3.2.2 Thinning:

The process in which dictation of features and shape classification are done. Freeman code Border is also used for this process.

In thinning process we can check the absolute position of pixels

3.3.3 Pre-Processing

The Arabic handwritten character is scanned off from the input device Tablet PC or digital pen with a resolution of 100 dpi and save it as a bmp format, The strokes (a mark made by digital pen, pencil, or paintbrush once across paper or canvas)These strokes consist of coordinates time like x,y etc.As every person writes with different force and angles and it depends on personal action and their tendency to move in various direction, so consider all the minor features of a input images for further processing. As in Arabic language has one or more strokes, the first one are main stroke (primary) and the other is secondary. Each character or sub character in Arabic can consisting of two more joint letters called ligature. These joint letters contains secondary and primary strokes. The binding of these joints letters creates different words. Each of which contains one primary stroke and 0 or more secondary. The other higher level technique used for feature extraction of online Arabic character recognition is put the characters in boxes or rectangle shape structure so that we can differentiate the characters more easily.

Firstly the proposed system uses putting the characters in boxes (Rectangle shape surrounded the characters) or making boundaries around the characters. The characters are places in the rectangle or box according to their shape and size. For this some are put in a single box and other are into two or more boxes or rectangle. For example the characters jeem, Aain Hhay are almost same size and similar shape so put these characters in same size box or rectangle. The character like Bay (), Fay() Tay() etc are almost same body but the difference is the position of dot objects are placed in same type boxes. The characters like seen (), swat (), etc are places into two boxes according to their size and shape

By putting the characters into rectangle like box helped to categories the same shape character in one place.becaues of this categorization is useful for higher level recognitions of character, it means differentiation among the characters will be easier. For making boundries around the characters or boxes we can use Matlab7.the code are here

3.3.1 Feature Extraction in Matlab

Matlab code for making boxes and sub boxes around the characters

This process fail to recognize the characters features more briefly so moves another technique called stroke identification, by the help of this a litter bit more feature of characters are retrieved , these features are more helpful in character recognition process using these features the proposed system try to recognize the online input character images.

Strokes identification is we can find the starting, ending and turning points of the characters , the characters contain one or more strokes, the first strokes is called base or primary strokes and the others is called secondary strokes. The rule is that first find the starting point of the characters, and called it a primary strokes, by using this starting point or staring pixel we can find the other stroke (secondary) and making secondary strokes as primary we can go for other strokes as well, the final strokes will be the ending points, for this we can find the starting and ending points (Pixels) of the character. Strokes process provides more information about the characters.

3.3.2 Classification of same shape characters

We can place the characters in a rectangle or box shape structure for the classification purpose. In technique use a Matlab and categories the character according o the body shape. For example the characters "bay" "tay" having same body structure and differentiated by the position of the dots objects. Which are above or the characters, similarly the bodies of characters jeem, Haa, Khaa are same, the only difference is the position of the dots objects. The characters Aain and Ghain are also differentiated by the dots.

So considering the dots objects and the body of the characters we can classify the characters. This is classification is helpful in the recognition process.

3.4 Feature Extraction

After preprocessing of the online characters images the next step is the main part which is the feature extraction stage, in this stage the retrieved images and their features like strokes and the rectangle boxes are treated using Matlab7, using Matlab we can code and find the features of characters and operate more operation on it.

The properties of characters that distinguish one from other are called feature extraction. These features or properties are used by user for recognition of the characters. two types of Features extraction are used one is relating to quantity or measure the quantity of features(Quantitative) and the 2nd is used to describing the size, appearance etc(qualitative) of the features of characters, The features like dots numbers, width ,height, area etc of the characters are in the quantitative features, and the characteristic like starting point ending points and dot position and other strokes properties are present in the types of features extraction called qualitative features

Different person writes the letters with different shapes like height, size is different from each others. Some become thick and some become thin so the thinning process is apply and try to make the characters in same size and shapes. Also check the density of the characters areas where the black and white pixels resides.

Strokes order: The movements of pen are detected from pixel to pixel for the direction of strokes as strokes are the movements of digital pen. in this we can find the starting point or pixel of characters and then moves from pixel to pixel to find the next stroke or turning point or pixel, some characters lines are crossed with each others. to find the cross point we can the neighbors of the pixel like the pixel contain more than two neighbors this pixel will be the crossing point of the characters lines. For this strokes identification we can find the numbers of joints, cross and others features as well like termination points

3.4.1Connected component construction:

Finding of connected component of online Arabic character is one of the preprocessing stages. In this stage the component are placed in the boxes region having 4-8 pixels which are connected with each other.

The process used to receive the connected component is a simple it reparative method which compares successive scan line of the data(Arabic character)which is used to determine the black pixels in any of line scan are connected together. In rectangles are the denoting a record that play longer than the most singles to enfold any grouping of connected black pixels between successive scan lines.

3.5 Recognition Based On Features Extraction

Firstly apply the boxes and rectangles techniques on the characters which differentiate more different characters with each others to reduce the execution times. We can also differentiate the characters with each others on the basis of secondary objects like dots, this can also reduce the execution times because this is easier to differentiate the characters with having and have not any dots.

In this process the retrieved the characteristics of the online data like characters are put to feed the system for recognizing purpose, this technique is called training. The character samples are provide to this training system

3.6 Results:

The ending portion of this work is to create a database of Extracted features of acquired data (online Arabic character) which is used in recognition process by comparing the original images which are already present in the hard drive.

All the process is of the online character recognition is done in Matlab. Like acquiring data, strokes identification recognition etc. using Matlab build in or self created function or m files features of characters are gained and put these features in training steps for recognition process. Different algorithms for used to find the features extraction and recognition process. In this work the simple neighbour technique are used. By using this technique on different features of the characters the following feature table are created.

The results of Arabic character recognition technique are stored in table of database for comparison of the original images which are already present in the database from sampling. The analysis of this result the accuracy is found in 70s % or more

Result obtained from AOCR process is stored in database. Those results are compared with the real image that exists in database that was gathered as input samples. On the bases of that result accuracy of a character is found which is approximately at least 60% to 80% or more


Chapter 4


For the efficient characters recognition the stroke identification is the most important stage, in this input as raw data is put in the system. Due to The complexities of Arabic characters features finding of online characters are not enough. In this proposed work focus on the preprocessing of online data in this case is Arabic characters; the feature extraction of these characters in this system is strokes identification and recognition stage. In this work use some build in algorithms to the features of characters and characters matching. So by strokes identification technique the efficiency of recognition is improved. This technique also monitors the person or writer activity like pen-movements pen up\down movements and time of pen movements. Database of Extracted features are created and compare theses features with raw data of handwritten character sample. By the help of this technique an efficient recognition of characters is obtained



  • Ismael Ahmad Jannoud "Automatic Arabic Hand Written Text Recognition System" Damascus University, Damascus, Syria and Al-zaytoonah University, Amman, Jordan.2007
  • Mohammed Z. Khedher, Gheith A. Abandah, and Ahmed M. Al-Khawaldeh" Optimizing Feature Selection for Recognizing Handwritten Arabic Characters" World Academy of Science, Engineering and Technology 4 2005
  • Rejean Plamondon and Sargur N.Srihari"online and offline Handwriting Recognition A comprehensive Survey"IEEE Transactions on pattern Analysis and machine intelligence vol 22.No 1 January 2000
  • Al-Taani Ahmad, Hammad Maen," Recognition of On-line Handwritten Arabic Digits Using Structural Features and Transition Network" September 12, 2006
  • Abdurazzag Ali ABURAS and Salem M. A. REHIEL" Off-line Omni-style Handwriting Arabic Character Recognition System Based on Wavelet Compression" 21 October 2007
  • Ayse Onat, Ferruh Yildiz, and Mesut Gündüz" Ottoman Script Recognition Using Hidden
  • Markov Model" World Academy of Science, Engineering and Technology 14 2006
  • Abdelmalek Zidouri, Muhammad Sarfraz" on optical character recognition of Arabic text" The 6th Saudi Engineering Conference, KFUPM, Dhahran, December 2002
  • Mohammed Z. Khedher, Gheith A. Abandah, and Ahmed M. Al-Khawaldeh" Optimizing Feature Selection for Recognizing Handwritten Arabic Characters" World Academy of Science, Engineering and Technology 4 2005
  • Abdurazzag Ali Aburas and Salem Ali Rehiel" JPEG for Arabic Handwritten Character Recognition: Add a Dimension of Application"
  • A. Amin. (2003). "Recognition of Hand-Printed Characters Based on Structural Description and Inductive Logic Programming" Pattern Recognition Letters, vol. 24, pp. 3187- 3196.
  • G. Abandah and F. Khundakjie, "Issues concerning code system for Arabic letters," Dirasat Engineering Sciences J., vol. 31, no. 1, pp 165-177, April 2004.
  • A. Amin, "Arabic character recognition," Handbook of Character Recognition and Document Image Analysis, pp 397-420, World Scientific, 1997.
  • Y. Al-Ohali, "Development and evaluation environment for typewritten Arabic character recognition" M.Sc. thesis, King Saud University, Saudi Arabia, 1995.
  • A. Saleh, "A method of coding hand-written Arabic characters and its Application to context-free grammar" Pattern Recognition Letters, vol. 15, pp. 1265-1271, Dec. 1994.
  • A. Amin, H. Al-sadouni, and S. Fischer, "Hand-printed Arabic character Recognition system using an artificial network," Pattern Recognition, vol.29, no.4, pp. 663-675, April 1996.
  • Namazi, M. and K. Faez "Recognition of multifont Farsi/Arabic characters using a fussy neural network," Proceedings of IEEE region 10 Annual International Conference vol. 2 pp. 918- 922 Nov. 1996
  • Zidouri, A., Chinveeraphan, S, and Sato, M "Structural Features by MCR Expression Applied to Printed Arabic Character Recognition" in 8th Int. Conf. on Image Analysis and Processing, (San Remo Italy), pp.557--562, Sept. 13-15 1995
  • C. C. Tappert, C. Y. Suen, and T. Wakahara." The state of the art in on-line handwriting recognition. IEEE Trans. On Pattern Analysis and Machine Intelligence, 12 ( 8), pp. 787-808, 1990.
  • R. G. Casey and E. Lecolinet". Strategies in character segmentation: A survey. Proceedings of International Conference on Document Analysis and Recognition" 1028-1033, 1995.