Gamal Tharwat

Deaf and dumb people struggle with communicating on a day-to-day basis. Current advancements in artificial intelligence (AI) have allowed this communication barrier to be removed. A letter recognition system for Arabic sign language (ArSL) has been developed as a result of this effort. The deep convolutional neural network (CNN) structure is used by the ArSL recognition system in order to process depth data and to improve the ability for hearing-impaired to communicate with others. In the proposed model, letters of the hand-sign alphabet and the Arabic alphabet would be recognized and identified automatically based on user input. The proposed model should be able to identify ArSL with a rate of accuracy of 97.1%. In order to test our approach, we carried out a comparative study and discovered that it is able to differentiate between static indications with a higher level of accuracy than prior studies had achieved using the same dataset.

Deaf and dumb Muslims cannot reach advanced levels of education due to the impact of obstruction on their educational attainment. This leads to their inability to learn, recite, and understand the meanings and interpretations of the Holy Qur’an as easily as ordinary people, which also prevents them from applying Islamic rituals such as prayer that require learning and reading the Holy Qur’an. In this paper, we propose a new model for Qur’anic sign language recognition based on convolutional neural networks through data preparation, preprocessing, feature extraction, and classification stages. The proposed model is aimed at recognizing the movements of the Arabic sign language by recognizing the hand gestures that refer to the dashed Qur’anic letters in order to help the deaf and dumb learn their Islamic rituals. The experiments have been conducted on a part of a large Arabic sign language dataset called ArSL2018, which represents the 14 dashed letters in the Holy Qur’an, so that this part contains only 24,137 images. The experimental results demonstrate that the proposed model performs better than the other existing models.

The COVID-19 pandemic has changed the lifestyle in global education systems in favor of more reliance on e-learning, which is believed to be a change that will continue. Exams are no exception as they are also moving towards online remote exams where examinees can take their exams at their remote locations. However, it is worth noting that it is still difficult to ensure that the full discipline of the exam session is administered remotely, as is the case for traditional exams that are conducted in the classroom and in the presence of proctors. However, there are many attempts at the commercial and research levels to overcome many of the challenges facing this problem from different angles. Therefore, this survey article attempts to review some of those attempts at both the solutions and technology levels. The article also made some comparisons and evaluations of the techniques and algorithms reviewed as an attempt to assess their weaknesses and strengths in counteracting persistent attempts to circumvent the discipline of the exam system. Hence, this article aims to pave the way for system designers who wish to implement remote exam session management systems by reviewing both commercially available and under-researched technologies. It also aims to introduce researchers to the findings of scientific research so far, while shedding light on open gaps and problems that have not yet been researched. غيّرت جائحة كورونا (COVID-19) نمط الحياة في أنظمة التعليم العالمية لصالح المزيد من الاعتماد على التعلم الإلكتروني، والذي يُعتقد أنه تغيير سيستمر. الامتحانات ليست استثناء حيث إنها تتجه أيضًا نحو الاختبارات عن بعد عبر الإنترنت حيث يمكن للممتحنين إجراء امتحاناتهم في مواقعهم البعيدة. ولكن يجدر ذكر انه لا يزال من الصعب ضمان إدارة الانضباط الكامل لجلسة الامتحان عن بعد كما هو الحال بالنسبة للامتحانات التقليديةالتي تتم في الفصول وبحضور المراقبين.إلا انه يوجد العديد منالمحاولات على المستويين التجاري والبحثي للتغلب على الكثير من التحدياتالتي تواجه هذه المشكلة من زوايا متفرقة. لذلك، تحاول مقالة الاستطلاع هذه مراجعة بعض تلك المحاولات على مستويي الحلول والتقنيات. وقد قام المقال أيضًا بإجراء بعض المقارنات والتقييمات للتقنيات والخوارزميات التي تمت مراجعتها كمحاولة...

E-Learning is become the education method of the future. Lecture videos are powerful and expressive learning resources that are commonly and extensively used in e-learning. However, one lecture video usually covers many topics/subtopics or even many conceptual instructional roles for a single topic. In adaptive e-Learning, the e-Learning of the future, requires small granular objects (Micro Learning Objects MLO) for more flexible and adaptive presentation of the lecture, hence, arose is the need for splitting this lecture video into many MLOs each playing a specific instructional role in the lecture. This research is concerned with the automatic extraction of MLOs out of existing lecture videos also with automatic annotation with the appropriate metadata attributes needed for the appropriate selection of the MLO by the adaptive process. This article, however, assumes that the lecture video is based on PowerPoint (or presentation) slides that are also available to the proposed MLO extraction process. أصبح التعلم اإللکتروني طريقة تعليم المستقبل. تعد مقاطع فيديو المحاضرات مصادر تعليمية قوية ومعبرة والتى يتم استخدامها بشکل شائع وعلى نطاق واسع في التعلم اإللکتروني. ومع ذلک ، عادةً ما يغطي مقطع فيديو محاضرة واحد العديد من الموضوعات / الموضوعات الفرعية أو حتى العديد من األدوار التعليمية المفاهيمية لموضوع واحد. في التعلم اإللکتروني التکيفي ، يتطلب التعلم اإللکتروني للمستقبل کائنات صغيرة )MLO Objects Learning Micro )لتقديم عرض أکثر مرونة وتکيفًا للمحاضرة ، ومن ثم نشأت الحاجة إلى تقسيم فيديو المحاضرة إلى العديد من MLOs .کل MLO له دور تعليمي محدد في المحاضرة. يهتم هذا البحث باالستخراج التلقائي لـ MLOs من مقاطع فيديو المحاضرات الحالية أي ًضا مع التعليق التوضيحي التلقائي بسمات البيانات الوصفية المناسبة الالزمة لالختيار المناسب لـ MLO من خالل العملية التکيفية. ومع ذلک ، تفترض هذه المقالة أن فيديو المحاضرة يستند إلى شرائح PowerPoint( أو العرض التقديمي( المتوفرة أي ًضا لعملية...

In recent years, the role of pattern recognition in systems based on human computer interaction (HCI) has spread in terms of computer vision applications and machine learning, and one of the most important of these applications is to recognize the hand gestures used in dealing with deaf people, in particular to recognize the dashed letters in surahs of the Quran. In this paper, we suggest an Arabic Alphabet Sign Language Recognition System (AArSLRS) using the vision-based approach. The proposed system consists of four stages: the stage of data processing, preprocessing of data, feature extraction, and classification. The system deals with three types of datasets: data dealing with bare hands and a dark background, data dealing with bare hands, but with a light background, and data dealing with hands wearing dark colored gloves. AArSLRS begins with obtaining an image of the alphabet gestures, then revealing the hand from the image and isolating it from the background using one of the proposed methods, after which the hand features are extracted according to the selection method used to extract them. Regarding the classification process in this system, we have used supervised learning techniques for the classification of 28-letter Arabic alphabet using 9240 images. We focused on the classification for 14 alphabetic letters that represent the first Quran surahs in the Quranic sign language (QSL). AArSLRS achieved an accuracy of 99.5 for the K-Nearest Neighbor (KNN) classifier.

Online learning has become at the core of educationsystems, especiallyafter the Corona pandemic. Noteworthy, LMS (Learning Management System) records all students' actions, whichisdifficult in traditionaleducationsystems,which, if properlyanalyzed, can give a clearpicture of the student'slearningbehavior. This studyfocuses on measuring and analyzingstudents’ engagement with courses, thusprovidingstudents, teachers, and academicadvisorswithrecommendations for improvement. Research has shownthat the more students’ engagement thehigher the success rate. Therefore, weconsiderhigh engagement as a key successindicatorin aneducation system. This researchidentifiedthe system-interaction attributesinfluencing the student’s engagementwith a course. It integratesthoseattributesinto a model for measuring and evaluating the degree of student’s engagement—the LAST-CASE Model. The LAST-CASE model isdesignedto measure the different engagement dimensions, such as cognitive, social, academic, and affective dimensions as recommended by the Community of Inquiry (CoI). This article describes how to measuresuchattributesusingboth the LMS log and the xAPImethod. In addition, we shed lights on how to assess the degree of “engagement” on differentthemes, e.g., for a student in a course, for a studentacross all courses in a particularfield, for differentstudents in different classes for the same course but taught by differentteachers, etc. Keywords: Student’ engagement; learning management system; theoretical framework; Adaptive e-Learning; recommendations; learning analytics; learner model. أصبح التعلم عبر الإنترنت في صميم أنظمة التعليم الحديثة، خاصة بعد جائحة کورونا. ومن الجدير بالذکر أن أنظمةادارة التعليمLMS تسجل جميع تفاعلات الطلاب مع النظام، وهو أمر لا يمکن القيام به في الأنظمة التقليدية، والذي إذا تم تحليله بشکل صحيح، يمکن أن يعطي صورة واضحة عن السلوک التعلمي للطالب. هذا البحث،يرکز على قياس وتحليل درجة "انخراط" الطالب (student engagement) في المقررات التي يدرسها، وبالتالي تقديم نصائح التحسين للطالب والمدرس والمرشد الأکاديمي.أثبتت الأبحاث أنه کلما زاد"انخراط" الطلاب في عملية التعلم زادت نسبة النجاح. لذلک، فإننا نعتبر "الانخراط" مؤشر نجاح رئيسي لنظام التعليم. حدد هذا البحث مجموعة من مؤشرات التفاعل مع النظامالمؤثرة في انخراط الطالب مع المقرر. ولقياس وتقييم درجة "انخراط الطالب تم التکامل بين هذه المؤشرات في نموذج LAST-CASE. تم تصميم نموذج...

This Gestures is one of the best ways of communication between dumb and blind people depend on the expression of signs. In this paper we suggest an algorithm to recognizing hand gestures of Arabic letters to communicate between the dumb (through signs) and blind (hear the voice corresponding to sings). The proposed algorithm used the video of gesture from the dumb then convert the video into frames ( images) and calculate the distance to recognition the letters by using k-mean, k- medoid and artificial neural network, calculate the distance by using Euclidean distance and slop. There are sixteen features (8-features from Euclidean distance and 8-features from slop ). The results were (93.3673% For k-mean),(93.1354% for k-medoid ) and(92.9499% for ANN). We create our data base (from 5- videos with 308 frames).