Abdelmalek Toumi - Academia.edu (original) (raw)

Papers by Abdelmalek Toumi

HAL (Le Centre pour la Communication Scientifique Directe), Nov 21, 2023

HAL (Le Centre pour la Communication Scientifique Directe), Jun 5, 2023

Global Navigation Satellite Systems (GNSS) are often the target of malicious attacks and interfer... more Global Navigation Satellite Systems (GNSS) are often the target of malicious attacks and interferences, mainly spoofing, thus posing a significant threat to both civilian and military equipment, and therefore necessitating effective detection and identification of such attacks. In this 'Work-in-Progress' paper, we propose the application of Machine Learning neural networks, a methodology proven highly effective in fields like cyberattack detection, to identify spoofing events across various scenarios. Our approach consists in computing non-time related metrics from a dataset of known spoofed signals, using the observables and signal-level measurements provided by a GNSS software receiver. The training is validated on both spoofed and clean scenarios to ensure a comprehensive approach. Furthermore, we provide a description of the feature's importance in the decision-making process of the model.

Automatic Target Recognition (ATR) is an interest problem in various application fields (security... more Automatic Target Recognition (ATR) is an interest problem in various application fields (security, surveillance, automotive, environment, medicine, communications, remote sensing, ...). Thus, SAR (Synthetic Aperture Radar) and ISAR (Inverse Synthetic Aperture Radar) radar images provide rich visual information about the observed radar target. From these radar images, several methods have been proposed to meet the expected requirements in several application domains, including target recognition, which is one of the main issues addressed in the present work. Traditional standard image classification techniques are not suitable for efficient classification of SAR images due to the limited data available in some classes (unbalanced data). To solve these problems, we introduce a deep learning model, the Siamese network with multiclass classification, built from a pre-trained model to improve the model performances on unbalanced classes. To evaluate the proposed method, the MSTAR dataset is used. The proposed method improves the recognition rate from 95,18% to 97.16%.

HAL (Le Centre pour la Communication Scientifique Directe), Jul 6, 2023

HAL is a multidisciplinary open access archive for the deposit and dissemination of scientific re... more HAL is a multidisciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L'archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d'enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

Advances in intelligent systems and computing, Jun 15, 2019

In this paper, we consider the problem of autonomous task allocation and trajectory planning for ... more In this paper, we consider the problem of autonomous task allocation and trajectory planning for a set of UAVs. This is a bi-level problem: the upper-level is a task assignment problem, subjected to UAV capability constraints; the lower-level constructs the detailed trajectory of UAVs, subjected to dynamics, avoidance and dependency constraints. Although the entire problem can be formulated as a mixed-integer linear program (MILP), and thus it can be solved by available software, the computational time increases intensively. For solving more efficiently this problem we propose an efficient approach based on the column generation method in which the modified dependency constraint will be added into the sub-problem. The performance of our approach is evaluated by comparing with solution given by the CPLEX on different scenarios.

HAL (Le Centre pour la Communication Scientifique Directe), Nov 24, 2010

HAL (Le Centre pour la Communication Scientifique Directe), 2010

ABSTRACT We describe in this paper, data processing algorithms applied on radar image in order to... more ABSTRACT We describe in this paper, data processing algorithms applied on radar image in order to extract feature descriptors and then to perform recognition task. Several kinds of descriptors can be used to acquire information about target characteristics from radar images such as ISAR (Inverse Synthetic Aperture Radar) images. This paper presents two types of vector descriptors extracted via two minds of transformed images so-called polar and log-polar images obtained respectively from the polar and log-polar mapping. In order to guarantee the invariance of some geometrical transformation, additional processing are proposed. In this paper, we present the polar and log-polar transformations and then the classification scheme adapted on correspondent polar and log-polar templates. In the classification step, log-polar and polar mapping results are compared using adapted classification scheme.

HAL (Le Centre pour la Communication Scientifique Directe), Jul 1, 2013

The target recognition from Radar images was a crucial step in our research. This paper presents ... more The target recognition from Radar images was a crucial step in our research. This paper presents a process a nd an adopted approach f or Automatic Target recognition using Inverse Synthetic Aperture Radar (ISAR) image . Indeed, the process adopted is composed of three steps. In the first step , we achieve the edge detection using of three techniques : Fisher, K - means and Expectation - M aximization (E - M) . Each of these techniques is combined with Watersheds (WS) algorithm to obtain the closed target shape. In order to ensure that the shape descriptors must be accurate, compact and invariant to several geometrical transformations (tran slation, rotation, scal e, etc.), we have used Fourier Descriptor computed on each obtained shape. To achieve a classification task in the last step, several techniques can be used to perform recognition tasks. We have used the nearest - neighbor classifier to retrieve a nearest kn own target for each unknow n target in the test dataset. Finally, in order to validate our proposed approach a database of ISAR images reconstructed from anechoic chamber simulations will be used . The simulation results using Fisher, E - M and K - Means method s will be presented in the last section of this paper

This paper presents a generic and versatile approach for implementing combining rules on preproce... more This paper presents a generic and versatile approach for implementing combining rules on preprocessed belief functions, issuing from a large population of information sources. In this paper, we address two issues, which are the intrinsic complexity of the rules processing, and the possible large amount of requested combinations. We present a fully distributed approach, based on a MapReduce scheme. This approach is generic. In particular, we compare two implementations of three sources Dubois & Prade rule within framework Apache Spark and Apache Flink.

2022 6th International Conference on Advanced Technologies for Signal and Image Processing (ATSIP)

Advances in Intelligent Systems and Computing, 2019

HAL (Le Centre pour la Communication Scientifique Directe), Nov 28, 2015

International audienceCe travail s’inscrit dans le cadre des travaux de recherche qui visent le t... more International audienceCe travail s’inscrit dans le cadre des travaux de recherche qui visent le traitement et l’exploitation des images radar dites ISAR(Inverse Sythetic Aperture Radar). Cette th´ematique de recherche est d’une grande importance dans diverses applications en environnementincertain a´erien et maritime. L’objectif phare concerne la proposition d’une chaˆıne compl`ete pour l’aide `a la reconnaissance automatique descibles `a partir des images ISAR. Dans cette optique, nous adoptons le processus d’extraction de connaissance `a partir de donn´ees (ECD) pourfournir des indicateurs d’aide `a la prise de d´ecision. Cette m´ethodologie est constitu´ee g´en´eralement de cinq phases allant de l’acquisition suiviede la la pr´eparation des donn´ees (pr´e-traitement des donn´ees et extraction des param`etres) jusqu’`a l’interpr´etation et l’´evaluation des r´esultats,en passant par la phase de fouille dans les donn´ees (datamining : classification et fusion). Dans cette communication, diff´erentes phases duprocessus sont pr´esent´ees. Ainsi, chaque brique est expos´ee avec les m´ethodes retenues dans le cadre de l’application radar abord´ee. Un attentionparticuli`ere est port´ee aux aspects op´erationnels en vue de la mise `a disposition d’un syst`eme pratique et optimis´e

HAL (Le Centre pour la Communication Scientifique Directe), Dec 23, 2016

International audienc

2020 5th International Conference on Advanced Technologies for Signal and Image Processing (ATSIP), 2020

This paper is devoted to the study of an automatic target recognition method to classify Inverse ... more This paper is devoted to the study of an automatic target recognition method to classify Inverse Synthetic Aperture Radar (ISAR) images of moving targets. 2-D ISAR imagery generation allows to obtain a pertinent representation of the moving target reflectivity distribution employing radar imaging system. The proposed target recognition technique is based on a combined the polar representation with shape matrix description applied on ISAR image. These descriptors (features) are used to achieve the recognition task method using Neural Network and Support Vector Machine. Several simulations are provided to validate the performances of the proposed method for the automatic aircraft target recognition.

2018 4th International Conference on Advanced Technologies for Signal and Image Processing (ATSIP), 2018

A new method for Automatic Radar Targets Recognition is presented based on Inverse Synthetic Aper... more A new method for Automatic Radar Targets Recognition is presented based on Inverse Synthetic Aperture Radar (ISAR). In this work, the first step is to construct ISAR images via a Non uniformly Sampled Bivariate Empirical Mode Decomposition Time-Frequency Distribution (NSBEMD-TFD) method. Indeed, this Time-Frequency representation is well suited for non-stationary signals analysis and provides high resolution with good accuracy. The obtained ISAR images is used to provide the evolution of two-dimensional spatial distribution of a moving target and, therefore, its are suitable to be used for radar target recognition tasks. In second step, a feature vectors are extracted from each ISAR images in order to describe the discriminative informations about a target. In the features extraction step, we computed several rings of polar space applied on ISAR image. Then, these rings is projected on 1-D vector. To ensure translation invariance of the obtained projected 1-D vector, a Fourier Descriptors are computed. In third step of this work, the recognition task is achieved using k-Nearest Neighbors (K-NN), Fuzzy k-NN, Neural network and Bayesian classifiers. To validate our approach, simulation results are presented on a set of several targets constituted by ideal point scatterers models.

IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium, 2018

This paper proposes a new target recognition method for inverse synthetic aperture radar (ISAR) i... more This paper proposes a new target recognition method for inverse synthetic aperture radar (ISAR) images. This method is based on joint statistical modeling of the complex wavelet coefficients for ISAR image characterization and the sparse representation based classification (SRC) for the recognition. To extract features from an ISAR image, we first transform it in the complex wavelet domain using the dual-tree complex wavelet transform (DT-CWT). Then, we compute magnitude information for each complex subband. After that, we propose a joint statistical model for magnitude distribution, that takes into account the dependences between different orientations and scales. To do so, we adopt the copula as a multivariate model thanks to its suitability to capture jointly the subband marginal distribution and the dependence structure. For the recognition step, we exploit SRC which recovers the test descriptor to classify over a given dictionary composed by the training descriptors. This method classifies the test sample as the class whose training samples can generate the minimum sparse representation error. Experimental results on ISAR images database show that using copula and sparse classifier improve significantly the recognition rates compared to classical models and classifiers.

IEEE Geoscience and Remote Sensing Letters, 2017

In this letter, we present a novel generic approach for radar automatic target recognition in eit... more In this letter, we present a novel generic approach for radar automatic target recognition in either inverse synthetic aperture radar (ISAR) or synthetic aperture radar (SAR) images. For this purpose, the radar image is described by a statistical modeling in the complex wavelet domain. Thus, the radar image is transformed into a complex wavelet domain using the dualtree complex wavelet transform. Afterward, the magnitudes of the complex sub-bands are modeled by Weibull or Gamma distributions. The estimated parameters of these models are stacked together to create a statistical dictionary in training step. For the recognition task, we use the weighted sparse representationbased classification method that captures the linearity and locality information of image features. In this context, we propose to use the Kullback-Leibler divergence between the parametric statistical models of training and test sets in order to assign a weight for each training sample. Experiments conducted on both ISAR and SAR images' databases demonstrate that the proposed approach leads to an improvement in the recognition rate.

Proceedings of the International Conference on Big Data and Advanced Wireless Technologies, 2016

Within the context of remote sensing and earth observation, technological development of radar sy... more Within the context of remote sensing and earth observation, technological development of radar systems is helping more and more researchers to exploit this domain, consequently this has given birth to a wide variety of applications in different fields. in the context of these applications, literature today others many studies and applications in order to solve inversion problems through experimental observations. In this paper we provide an original inversion method which consists in integrating the SPM electromagnetic scattering model into image processing in order to extract physical or geometric properties of a natural observed surface from a radar synthetic aperture radar image (SAR) acquired by TerraSAR-X observation satellite.