Abdurazzag Ali Aburas

Arabic handwriting recognition: Challenges and solutions

Optical Characters Recognition (OCR) is one of the active subjects of research since the early days of computer science. Even if Arabic characters are used by more than a half a billion people; Arabic characters recognition has not received enough interests by the researchers. Little research progress has been achieved comparing to what has been done with Latin and Chinese. The cursive nature of the Arabic characters makes it more difficult to achieve a high accuracy in character recognition since even printed Arabic characters are in cursive form. This paper presents the main challenges (difficulties) researchers are facing and up to dated solutions (the common methods) are used for Arabic text recognition.

Fuzzy Logic based algorithm for uniprocessor scheduling

The key to multiprogramming is scheduling. Multiprogramming enables an operating system to use its resources more efficiently. When the system has a range of processes to execute, it must also have a policy to decide which process to run at a given time. This policy should attempt to maximize throughput, minimize latency, prevent process starvation etc. There are many techniques that are nowadays in use. In this paper we present a new improve scheduling algorithm technique based on Fuzzy Logic. We use the Matlab implementation of this algorithm to compare and contrast the output with a currently used technique (HRRN). In this paper present a short overview of some of the well known techniques of short-term scheduling in a single processor environment. The performance of the new process scheduling algorithm based fuzzy logic shows that the batch processorspsila utilization could be increased and jobspsila average lateness with respect to the internal deadline reduced significantly by applying our methodology.

New Promising Off Line Tool for Arabic Handwritten Character Recognition Based On JPEG2000 Image Compression

The typical optical character recognition (OCR) systems, regardless the characters nature, are based mainly on three stages, preprocessing, features extraction and discrimination (recognizer). Each stage has its own problems and effects on the system efficiency such as time consuming and recognition errors. In order to avoid these difficulties this research paper presents new construction of OCR system for handwriting Arabic characters using JPEG2000 standard. The proposed algorithm obtained promising results in terms of accuracy as well as in terms of time consuming.

On-line homeschooling learning management system using intelligent tutoring system

Homeschooling is an effective tool for overcoming hinders and barriers that constrain the ability of a societys educational system to cover and reach students (or potential students) whom are living in suburban or distant areas. Currently, homeschooling is considered one of new type of education and increasingly important of less expensive type of study, especially for developing countries. Homeschooling can help societies attain better educational levels without having to spend huge funds on the construction and maintenance of traditional schools. This research work is introducing a new management system to improve the evaluation of homeschooling examinations. The new system shows better exams results and diagnoses of student feedback. The study aims to exploring the prospects of implementing the Intelligent Tutoring Systems — ITS in Libya and integrating ITS into the Libyan system of homeschooling program.

Fuzzy Logic Based Algorithm for Disk Scheduling Policy

Hard disk is the important component in computer system as data storage device. It is being used to store large amount of information in all modern computers. Disk performance can be improved by incorporating disk scheduling optimization for satisfy pending request in the disk queue. Operating system, being the interface between the user and the hardware, has the responsibility to use hardware efficiently. Current algorithm have not knowledge about the additional information provided by operating system and not been used for improvement of the seeking and/or tracking accessing. The paper present fuzzy model knowledge based for improving the disk scheduling. We use the Matlab implementation of this algorithm to compare and contrast the output with existing algorithms. Incorporating knowledge-based scheduling policy using fuzzy logic improves overall system behavior. The performance of the new disk scheduling algorithm based on fuzzy logic shows that the seek distance movement could be decreased by applying our methodology.

JPEG for Arabic Handwritten Character Recognition: Add a Dimension of Application

The ultimate objective of any Optical Character Recognition (OCR) system is to simulate the human reading capabilities. That is why OCR systems are considered a branch of artificial intelligence and a branch of computer vision as well [Sunji Mori (1999)] Character recognition has received a lot of attention and success for Latin and Chinese based languages, but this is not the case for Arabic and Arabic-like languages such as Urdu, Persian, Jawi, Pishtu and others [Abdelmalek Z (2004)]. Researchers classify OCR problem into two domains. One deals with the image of the character after it is input to the system by, for instant, scanning in which is called Off-line recognition. The other has different input way, where the writer writes directly to the system using, for example, light pen as a tool of input. This is called On-line recognition. The online problem is usually easier than the offline problem since more information is available [Liana M & Venu G (2006)]. These two domains (offline & online) can be further divided into two areas according to the character itself that is either handwritten or printed character. Roughly, the OCR system based on three main stages: preprocessing, feature extraction, and discrimination (called also, classifier, or recognition engine) Figure 1.1 depicts the block diagram of the typical OCR system. Traditional OCR systems are suffering from two main problems, one comes from features extraction stage and the other comes from classifier (recognition stage). Feature extraction stage is responsible for extracting features from the image and passing them as global or local information to the next stage in order to help the later taking decision and recognizing the character. Two challenges are faced; if feature extractor extracts many features in order to offer enough information for classifier, this means many computations as well as more complex algorithms are needed. Thus, long processor time will be consumed. On the other hand, if few features are extracted in order to speed up the process, insufficient information may be passed to classifier. The second main problem that classifier is responsible for, is that most of classifiers are based on Artificial Neural Networks (ANNs). However, to improve the intelligence of these ANNs, huge iterations, complex computations, and learning algorithms are needed, which also lead to consume the processor time. Therefore, if the recognition accuracy is improved, the consumed time will increase and vice versa. O pe n A cc es s D at ab as e w w w .ite ch on lin e. co m

Comprehensive review for Arabic handwriting and printed characters recognition

The characteristics of the Arabic character (handwritten or printed) make great challenge in the field of automatic optical character recognition OCR in terms of accuracy and recognition rate. This paper provides a comprehensive review of the latest techniques and methods have been proposed and their results. The characteristics of the handwriting Arabic characters are also discussed in more detail. Finally, a new method based on image compression is proposed to overcome the traditional OCRpsilas problems as well as to achieve both higher accuracy and higher rate.

Handwriting recognition system using fast wavelets transform

Optical Characters Recognition (OCR) is one of the active subjects of research since the early days of computer science. There are two main stages in most of OCR systems: features extraction and classification. Artificial Neural Networks and Hidden Markov Models are the most popular classification methods used for OCR systems. In this paper, a method that relays on Fast Wavelets Transform (FWT) for optical character recognition is proposed. The idea of the proposed technique is to use the FWT to produce a coefficient vector of the character images, which will be directly used to recognize characters. Using the proposed technique, an accuracy of 94.18% in average was achieved.

Knowledge-based disk scheduling policy using fuzzy logic

Hard disk is the important component in computer system as data storage device. It is being used to store large amount of information in all modern computers. However, hard drive speed improvement slower than their capacity. Thus, long service delay for I/O bound processes may occur. Disk scheduling is an important issue for providing a quick response time in query processing. Operating System being the interface between the user and the hardware has the responsibility to use hardware efficiently. One crucial branch of this function is to decrease the response time and allow for better bandwidth through disk scheduling. Disk scheduling is used when processes running on the machine have multiple requests for data from the disks and a particular order is needed to access them most effectively. There are two major operations in the disk scheduling head-positioning of a hard disk i.e. seeking and tracking. Current algorithm have not included the system knowledge about the additional information provided by Operating System and not been used for improvement of the seeking and/or tracking accessing. This research project applies fuzzy model knowledge based on improving the disk scheduling algorithm. Incorporating knowledge-based scheduling policy using fuzzy logic improves the average response time, average seeks distance and average seeks time.

Fuzzy based technique for microchip lead inspection using machine vision

This research develops a fuzzy based algorithm for microchip image lead Inspection. Leads are inspected for count, planarity, offset, pitch and span defects. Firstly, it utilizes fast preprocessing techniques and blobspsila features extraction methods to achieve high inspection rates. Using the features extracted, the algorithm first finds a proper thresholding value. It then applies fuzzy logic to make a decision on the status of the IC based on these features. The algorithm proposes a structured way for building the fuzzy systems as well as the associated set of inference rules.