abdelouahab attia - Academia.edu (original) (raw)

Papers by abdelouahab attia

Medical care has always presented quite wide ranged and challenging problems. However, machine le... more Medical care has always presented quite wide ranged and challenging problems. However, machine learning techniques and methods as well as deep learning never stopped evolving and tackling those challenges issued by medicine, medical and health care. In order to have a more close up look on how machine learning and deep learning has been affecting medical care in general, we review in this paper some machine learning and deep learning techniques used in a variety of medical care sections such as medical imaging, medical decision, diagnostic, medical records and big data, and disease prediction.

Diabetic Retinopathy (DR) is one of the mainly causes of visual loss worldwide. In fact, DR is le... more Diabetic Retinopathy (DR) is one of the mainly causes of visual loss worldwide. In fact, DR is leading source of impaired vision in people between 25 and 74 years old. DR exists in wide ranged and its detection is a challenging problem. The gradual deterioration of retina leads to DR with several types of lesions, including hemorrhages, exudates, micro aneurysms, etc. Early detection and diagnosis can prevent and save the vision of diabetic patients or at least the progression of DR can be slowed down. The manual diagnosis and analysis of fundus images to substantiate morphological changes in micro aneurysms, exudates, blood vessels, hemorrhages, and macula are usually time-consuming and monotonous task. It can be made easy and fast with the help of computer-aided system based on advanced machine learning techniques that can greatly help doctors and medical practitioners. Thus, the main focus of this paper is to provide a summary of the numerous methods designed for discovering hemo...

2018 International Conference on Communications and Electrical Engineering (ICCEE), 2018

among the numerous biometric systems existing in the literature, face identification systems have... more among the numerous biometric systems existing in the literature, face identification systems have received a considerable interest in latest years. This paper presents a novel approach to face-feature extraction based on the Adaptive Single scale Retinex algorithm (ASSR) and the Gabor filter-bank. The ASSR has been used to extract the illumination (I-image) and the reflectance images (R-image) from each original face image. NIimage (normalization illumination image) has been obtained by eliminating the uneven lighting from the I-image using morphological operations. Then, the Gabor filter bank is applied on the NI-image and the reflectance images to extract feature vectors of these images. These features have been concatenated to make a huge feature vector of every user. While in the next step PCA + LDA technique has been employed to reduce the dimensionality of these novel feature vectors of every user and to further improve its discriminatory power. Finally, the nearest neighbor c...

Biometric based face recognition is a successful method for automatically identifying a person us... more Biometric based face recognition is a successful method for automatically identifying a person using her face, with a high confidence. For that reason, this paper introduces an efficient method for face recognition based on deep networks. It considers the three face regions: eye, mouth, and face. First, we have built one sparse autoencoder for every single region their outputs will be concatenated together and fed into another sparse autoencoder. After that, the softmax layer has been employed in the classification step. However, with a deep network method known as the softmax layer has been formed by stacking the encoders from the autoencoder. Followed by formed the full deep network. Finally, the results have been generated on the test set based on the deep network. In the experimental stage, the Yale B database and the AR database and JAFFE database have been used to test the proposed individual recognition system. Experimental findings have clearly proven that the performance o...

KSII Transactions on Internet and Information Systems

Blind Source Separation (BSS) is a technique used to separate supposed independent sources of sig... more Blind Source Separation (BSS) is a technique used to separate supposed independent sources of signals from a given set of observations. In this paper, the High Exploration Particle Swarm Optimization (HEPSO) algorithm, which is an enhancement of the Particle Swarm Optimization (PSO) algorithm, has been used to separate a set of source signals. Compared to PSO algorithm, HEPSO algorithm depends on two additional operators. The first operator is based on the multi-crossover mechanism of the genetic algorithm while the second one relies on the bee colony mechanism. Both operators have been employed to update the velocity and the position of the particles respectively. Thus, they are used to find the optimal separating matrix. The proposed method enhances the overall efficiency of the standard PSO in terms of good exploration and performance. Based on many tests realized on speech and music signals supplied by the BSS demo, experimental results confirm the robustness and the accuracy of the introduced BSS technique.

Optical character recognition systems for handwritten Arabic language still face challenges, owin... more Optical character recognition systems for handwritten Arabic language still face challenges, owing to high level of ambiguity, complexity and tremendous variations in human writing styles. In this paper, we propose a new and effective Arabic handwritten characters recognition framework using multi-scale histogram oriented gradient (HOG) features and the deep rule-based classifier (DRB). In the feature extraction stage, the proposed framework combines multi-scale HOG features, and then the DRB is applied on comprehensive HOG features to obtain the final classification label/class. This study involves experimental analyses that were conducted on the publicly available cursive Arabic Handwritten Characters Database (AHCD) containing 16800 characters. Experimental results demonstrate the efficacy of the proposed recognition system compared to the existing state-of-the-art-

2020 4th International Symposium on Informatics and its Applications (ISIA)

Evolving Systems, Dec 19, 2019

Identifying epilepsy cases and epileptic seizures from electroencephalogram (EEG) signals is a ch... more Identifying epilepsy cases and epileptic seizures from electroencephalogram (EEG) signals is a challenging issue, which usually needs high level of skilled neurophysiologists. Numerous works have attempted to develop tools that can provide an assistant to neurophysiologist in analyzing the EEG for epileptic seizures detection. This paper proposes a new automatic framework to identify and classify the epileptic seizure from EEG using a machine learning method. In particular, the feature extraction process of the proposed scheme utilizes autoregressive model (AR) and firefly optimization (FA) to procure an optimal model order (P). Namely, the main aim of FA is to find the best model order (P) with minimum residual variance using Akaike information criterion (AIC) as an objective function of FA algorithm. A support vector machine (SVM) classifier is employed for the classification of the epileptic seizures signals. The presented scheme is also effective for short segment of EEG signals owing to use of AR model in features extraction stage. Experiments with the publicly available Bonn database that is composed of healthy (nonepileptic), interictal and ictal EEG samples show promising results with high accuracy.

Journal of Medical Imaging and Health Informatics

International Journal of Artificial Intelligence and Machine Learning

The optical character recognition (OCR) system is still an active research field in pattern recog... more The optical character recognition (OCR) system is still an active research field in pattern recognition. Such systems can identify, recognize and distinguish electronically between characters and texts, printed or handwritten. They can also do a transformation of such data type into machine-processable form to facilitate the interaction between user and machine in various applications. In this paper, we present the global structure of an OCR system, with its types (on-line and off-line), categories (printed and handwritten) and its main steps. We also focused on off-line handwritten Arabic character recognition and provided a list of the main datasets publicly available. This paper also presents a survey of the works that have been carried out over recent years. Finally, some open issues and potential research directions have been highlighted

Signal, Image and Video Processing

Evolving Systems

ICTACT Journal on Image and Video Processing

In this paper, a new method based on Log Gabor-TPLBP (LGTPLBP) has been proposed. However the Thr... more In this paper, a new method based on Log Gabor-TPLBP (LGTPLBP) has been proposed. However the Three Patch Local Binary Patterns (TPLBP) technique used in face recognition has been applied in Finger-Knuckle-Print (FKP) recognition. The 1D-Log Gabor filter has been used to extract the real and the imaginary images from each of the Region of Interest (ROI) of FKP images. Then the TPLBP descriptor on both images has been applied to extract the feature vectors of the real image and the imaginary image respectively. These feature vectors have been jointed to form a large feature vector for each image FKP. After that, the obtained feature vectors of all images are processed directly with a dimensionality reduction algorithm, using linear discriminant analysis (LDA). Finally, the cosine Mahalanobis distance (MAH) has been used for matching stage. To evaluate the effectiveness of the proposed system several experiments have been carried out. The Hong Kong Polytechnic University (PolyU) FKP database has been used during all of the tests. Experimental results show that the introduced system achieves better results than other stateof-the-art systems for both verification and identification.

Journal of Electronic Imaging

2016 8th International Conference on Modelling, Identification and Control (ICMIC), 2016

International Journal of Advanced Computer Science and Applications, 2016

In the functional Magnetic Resonance Imaging (fMRI) data analysis, detecting the activated voxels... more In the functional Magnetic Resonance Imaging (fMRI) data analysis, detecting the activated voxels is a challenging research problem where the existing methods have shown some limits. We propose a new method wherein brain mapping is done based on Dempster-Shafer theory of evidence (DS) that is a useful method in uncertain representation analysis. Dempster-Shafer allows finding the activated regions by checking the activated voxels in fMRI data. The activated brain areas related to a given stimulus are detected by using a belief measure as a metric for evaluating activated voxels. To test the performance of the proposed method, artificial and real auditory data have been employed. The comparison of the introduced method with the t-test and GLM method has clearly shown that the proposed method can provide a higher correct detection of activated voxels.