Yaser Saleh
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Yaser Saleh has a total of 8 co-authors. Here are the top co-authors by number of shared publications:
Eran Edirisinghe
Department Computer Science, Loughborough University, Loughborough, LE11 3TU, UK
2 shared publications- Ghassan Issa
Faculty of Information Technology, University of Petra, Amman, Jordan
1 shared publication - Izzeddin Matar
University of Petra
1 shared publication - Mohammad AlShaikh Hasan
Brunel University, UK
1 shared publication - Muhanna Muhanna
Princess Sumaya University for Technology
1 shared publication - Nesreen Otoum
Department of Software Engineering, University of Petra, Amman, Jordan
1 shared publication
Latest Publications
- Open Access
- 13 June 2023
- Published by International Association of Online Engineering (IAOE) in International Journal of Online and Biomedical Engineering (iJOE)
- Vol. 19 (07) , 26-38
Archeological Sites Classification Through Partial Imaging and Convolutional Neural Networks
In this paper, a novel approach for classifying archeological sites using publicly available images through the use of Convolutional Neural Networks (CNNs) is presented. To surmount the problem of having a limited amount of data to use in training and testing the CNNs, our approach employs the technique of fine tuning. We conducted an experiment with four popular CNN architectures: VGG-16, VGG-19, ResNet50, and InceptionV3. The results show that our models achieved an impressive accuracy of up to 98% using the VGG-16 and InceptionV3 models and up to 97% using the ResNet50 model, while the VGG-19 model produced results with an accuracy of 95%. The results of this study demonstrate the effectiveness of our proposed approach in classifying archeological sites using publicly available images and highlight the potential of deep learning techniques for archeological site classification.
- 13 June 2023
- Published by International Association of Online Engineering (IAOE) in International Journal of Online and Biomedical Engineering (iJOE)
- Vol. 19 (07) , 26-38
Open AccessArcheological Sites Classification Through Partial Imaging and Convolutional Neural Networks
In this paper, a novel approach for classifying archeological sites using publicly available images through the use of Convolutional Neural Networks (CNNs) is presented. To surmount the problem of having a limited amount of data to use in training and testing the CNNs, our approach employs the technique of fine tuning. We conducted an experiment with four popular CNN architectures: VGG-16, VGG-19, ResNet50, and InceptionV3. The results show that our models achieved an impressive accuracy of up to 98% using the VGG-16 and InceptionV3 models and up to 97% using the ResNet50 model, while the VGG-19 model produced results with an accuracy of 95%. The results of this study demonstrate the effectiveness of our proposed approach in classifying archeological sites using publicly available images and highlight the potential of deep learning techniques for archeological site classification.
- Open Access
- 16 September 2022
- Published by Association for Computing Machinery (ACM)
- p. 207-212
E-learning Quality Assurance Guidelines for Higher Education Institutions in Jordan
The Covid-19 pandemic affected many areas around the world, and with the lack of a vaccine, social distancing and keeping a high level of hygiene was the only actions people can rely on. Because of this change in the way of life, education was affected where it became not normal for students to socialize and have direct contact with teachers. Taking the case of higher education in Jordan, and its conversion into an online based learning approach, this paper examines an assessment of the quality of education that was taking place during the lock down. Then, it presents a literature study of guidelines published about quality of e-learning, and after evaluating the literature and the specific needs of Jordanian institutes, the paper presents a set of guidelines as a scorecard for higher education-al institutes to follow and maintain a high level of education quality in their online teaching approach.
- 16 September 2022
- Published by Association for Computing Machinery (ACM)
- p. 207-212
Open AccessE-learning Quality Assurance Guidelines for Higher Education Institutions in Jordan
The Covid-19 pandemic affected many areas around the world, and with the lack of a vaccine, social distancing and keeping a high level of hygiene was the only actions people can rely on. Because of this change in the way of life, education was affected where it became not normal for students to socialize and have direct contact with teachers. Taking the case of higher education in Jordan, and its conversion into an online based learning approach, this paper examines an assessment of the quality of education that was taking place during the lock down. Then, it presents a literature study of guidelines published about quality of e-learning, and after evaluating the literature and the specific needs of Jordanian institutes, the paper presents a set of guidelines as a scorecard for higher education-al institutes to follow and maintain a high level of education quality in their online teaching approach.
- Open Access
- 14 May 2020
- Published by International Association of Online Engineering (IAOE) in International Journal of Online and Biomedical Engineering (iJOE)
- Vol. 16 (05) , 71-83
Arabic Sign Language Recognition through Deep Neural Networks Fine-Tuning
Sign Language is considered the main communication tool for deaf or hearing impaired people. It is a visual language that uses hands and other parts of the body to provide people who are in need to full access of communication with the world. Accordingly, the automation of sign language recognition has become one of the important applications in the areas of Artificial Intelligence and Machine learning. Specifically speaking, Arabic sign language recognition has been studied and applied using various intelligent and traditional approaches, but with few attempts to improve the process using deep learning networks. This paper utilizes transfer learning and fine tuning deep convolutional neural networks (CNN) to improve the accuracy of recognizing 32 hand gestures from the Arabic sign language. The proposed methodology works by creating models matching the VGG16 and the ResNet152 structures, then, the pre-trained model weights are loaded into the layers of each network, and finally, our own soft-max classification layer is added as the final layer after the last fully connected layer. The networks were fed with normal 2D images of the different Arabic Sign Language data, and was able to provide accuracy of nearly 99%.
- 14 May 2020
- Published by International Association of Online Engineering (IAOE) in International Journal of Online and Biomedical Engineering (iJOE)
- Vol. 16 (05) , 71-83
Open AccessArabic Sign Language Recognition through Deep Neural Networks Fine-Tuning
Sign Language is considered the main communication tool for deaf or hearing impaired people. It is a visual language that uses hands and other parts of the body to provide people who are in need to full access of communication with the world. Accordingly, the automation of sign language recognition has become one of the important applications in the areas of Artificial Intelligence and Machine learning. Specifically speaking, Arabic sign language recognition has been studied and applied using various intelligent and traditional approaches, but with few attempts to improve the process using deep learning networks. This paper utilizes transfer learning and fine tuning deep convolutional neural networks (CNN) to improve the accuracy of recognizing 32 hand gestures from the Arabic sign language. The proposed methodology works by creating models matching the VGG16 and the ResNet152 structures, then, the pre-trained model weights are loaded into the layers of each network, and finally, our own soft-max classification layer is added as the final layer after the last fully connected layer. The networks were fed with normal 2D images of the different Arabic Sign Language data, and was able to provide accuracy of nearly 99%.
- 1 April 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 116-119
3D face reconstruction and recognition using the overfeat network
Although face recognition is considered a popular area of research and study, it still has few unresolved challenges, and with the appearance of devices such as the Microsoft Kinect, new possibilities for researchers were uncovered. With the goal of enhancing face recognition techniques, this paper presents a novel way to reconstruct face images in different angles, through the use of the data of one front image captured by the Kinect, using faster techniques than ever before, also, this paper utilizes a deep learning network called Overfeat, where it functioned as a feature extractor that was used on normal images and on the new 3D created images, which introduced a new application for the network. To check the capabilities of the new created images, they were used as a testing set in three main experiments. Finally, results of the experiments are presented to prove the ability of the created images to function as new data sets for face recognition; also, proving the capability of the Overfeat network, working with computer generated face images.
- 1 April 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 116-119
3D face reconstruction and recognition using the overfeat network
Although face recognition is considered a popular area of research and study, it still has few unresolved challenges, and with the appearance of devices such as the Microsoft Kinect, new possibilities for researchers were uncovered. With the goal of enhancing face recognition techniques, this paper presents a novel way to reconstruct face images in different angles, through the use of the data of one front image captured by the Kinect, using faster techniques than ever before, also, this paper utilizes a deep learning network called Overfeat, where it functioned as a feature extractor that was used on normal images and on the new 3D created images, which introduced a new application for the network. To check the capabilities of the new created images, they were used as a testing set in three main experiments. Finally, results of the experiments are presented to prove the ability of the created images to function as new data sets for face recognition; also, proving the capability of the Overfeat network, working with computer generated face images.
- Open Access
- 1 May 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
Novel approach to enhance face recognition using depth maps
Face recognition, although being a popular area of research and study, still has many challenges, and with the appearance of the Microsoft Kinect device, new possibilities of research were uncovered, one of which is face recognition using the Kinect. With the goal of enhancing face recognition, this paper is aiming to prove how depth maps, since not effected by illumination, can improve face recognition with a benchmark algorithm based on the Eigenface. This required some experiments to be carried out, mainly in order to check if algorithms created to recognize faces using normal images can be as effective if not more effective with depth map images. The OpenCV Eigenface algorithm implementation was used for the purpose of training and testing both normal and depth-map images. Finally, results of the experiments are presented to prove the ability of the tested algorithm to function with depth maps, also, proving the capability of depth maps face recognition's task in poor illumination.
- 1 May 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
Open AccessNovel approach to enhance face recognition using depth maps
Face recognition, although being a popular area of research and study, still has many challenges, and with the appearance of the Microsoft Kinect device, new possibilities of research were uncovered, one of which is face recognition using the Kinect. With the goal of enhancing face recognition, this paper is aiming to prove how depth maps, since not effected by illumination, can improve face recognition with a benchmark algorithm based on the Eigenface. This required some experiments to be carried out, mainly in order to check if algorithms created to recognize faces using normal images can be as effective if not more effective with depth map images. The OpenCV Eigenface algorithm implementation was used for the purpose of training and testing both normal and depth-map images. Finally, results of the experiments are presented to prove the ability of the tested algorithm to function with depth maps, also, proving the capability of depth maps face recognition's task in poor illumination.