Accurate segmentation of the lungs in CXR images is the basis for an automated CXR image analysis... more Accurate segmentation of the lungs in CXR images is the basis for an automated CXR image analysis system. It helps radiologists in detecting lung areas, subtle signs of disease and improving the diagnosis process for patients. However, precise semantic segmentation of lungs is considered a challenging case due to the presence of the edge rib cage, wide variation of lung shape, and lungs affected by diseases. In this paper, we address the problem of lung segmentation in healthy and unhealthy CXR images. Five models were developed and used in detecting and segmenting lung regions. Two loss functions and three benchmark datasets were employed to evaluate these models. Experimental results showed that the proposed models were able to extract salient global and local features from the input CXR images. The best performing model achieved an F1 score of 97.47%, outperforming recent published models. They proved their ability to separate lung regions from the rib cage and clavicle edges and segment varying lung shape depending on age and gender, as well as challenging cases of lungs affected by anomalies such as tuberculosis and the presence of nodules.
The simultaneous advances in deep learning and the Internet of Things (IoT) have benefited distri... more The simultaneous advances in deep learning and the Internet of Things (IoT) have benefited distributed deep learning paradigms. Federated learning is one of the most promising frameworks, where a server works with local learners to train a global model. The intrinsic heterogeneity of IoT devices, or non-independent and identically distributed (Non-I.I.D.) data, combined with the unstable communication network environment, causes a bottleneck that slows convergence and degrades learning efficiency. Additionally, the majority of weight averaging-based model aggregation approaches raise questions about learning fairness. In this paper, we propose a peer-to-peer federated learning (P2PFL) framework based on Vision Transformers (ViT) models to help solve some of the above issues and classify COVID-19 vs. normal cases on Chest-X-Ray (CXR) images. Particularly, clients jointly iterate and aggregate the models in order to build a robust model. The experimental results demonstrate that the p...
X-ray images are the most widely used medical imaging modality. They are affordable, non-dangerou... more X-ray images are the most widely used medical imaging modality. They are affordable, non-dangerous, accessible, and can be used to identify different diseases. Multiple computer-aided detection (CAD) systems using deep learning (DL) algorithms were recently proposed to support radiologists in identifying different diseases on medical images. In this paper, we propose a novel two-step approach for chest disease classification. The first is a multi-class classification step based on classifying X-ray images by infected organs into three classes (normal, lung disease, and heart disease). The second step of our approach is a binary classification of seven specific lungs and heart diseases. We use a consolidated dataset of 26,316 chest X-ray (CXR) images. Two deep learning methods are proposed in this paper. The first is called DC-ChestNet. It is based on ensembling deep convolutional neural network (DCNN) models. The second is named VT-ChestNet. It is based on a modified transformer model. VT-ChestNet achieved the best performance overcoming DC-ChestNet and state-of-the-art models (DenseNet121, DenseNet201, EfficientNetB5, and Xception). VT-ChestNet obtained an area under curve (AUC) of 95.13% for the first step. For the second step, it obtained an average AUC of 99.26% for heart diseases and an average AUC of 99.57% for lung diseases.
Wildland fires are one of the most dangerous natural risks, causing significant economic damage a... more Wildland fires are one of the most dangerous natural risks, causing significant economic damage and loss of lives worldwide. Every year, millions of hectares are lost, and experts warn that the frequency and severity of wildfires will increase in the coming years due to climate change. To mitigate these hazards, numerous deep learning models were developed to detect and map wildland fires, estimate their severity, and predict their spread. In this paper, we provide a comprehensive review of recent deep learning techniques for detecting, mapping, and predicting wildland fires using satellite remote sensing data. We begin by introducing remote sensing satellite systems and their use in wildfire monitoring. Next, we review the deep learning methods employed for these tasks, including fire detection and mapping, severity estimation, and spread prediction. We further present the popular datasets used in these studies. Finally, we address the challenges faced by these models to accurately...
The COVID-19 virus has made a huge impact on people’s lives ever since the outbreak happened in D... more The COVID-19 virus has made a huge impact on people’s lives ever since the outbreak happened in December 2019. Unfortunately, the COVID-19 virus has not completely vanished from the world yet, and thus, global agitation is still increasing with mutations and variants of the same. Early diagnosis is the best way to decline the mortality risk associated with it. This urges the necessity of developing new computational approaches that can analyze a large dataset and predict the disease in time. Currently, automated virus diagnosis is a major area of research for accurate and timely predictions. Artificial intelligent (AI)-based techniques such as machine learning (ML) and deep learning (DL) can be deployed for this purpose. In this, compared to traditional machine learning techniques, deep Learning approaches show prominent results. Yet it still requires optimization in terms of complex space problems. To address this issue, the proposed method combines deep learning predictive models ...
The world has seen an increase in the number of wildland fires in recent years due to various fac... more The world has seen an increase in the number of wildland fires in recent years due to various factors. Experts warn that the number of wildland fires will continue to increase in the coming years, mainly because of climate change. Numerous safety mechanisms such as remote fire detection systems based on deep learning models and vision transformers have been developed recently, showing promising solutions for these tasks. To the best of our knowledge, there are a limited number of published studies in the literature, which address the implementation of deep learning models for wildland fire classification, detection, and segmentation tasks. As such, in this paper, we present an up-to-date and comprehensive review and analysis of these vision methods and their performances. First, previous works related to wildland fire classification, detection, and segmentation based on deep learning including vision transformers are reviewed. Then, the most popular and public datasets used for thes...
To proactively mitigate malware threats, cybersecurity tools, such as anti-virus and anti-malware... more To proactively mitigate malware threats, cybersecurity tools, such as anti-virus and anti-malware software, as well as firewalls, require frequent updates and proactive implementation. However, processing the vast amounts of dataset examples can be overwhelming when relying solely on traditional methods. In cybersecurity workflows, recent advances in natural language processing (NLP) models can aid in proactively detecting various threats. In this paper, we present a novel approach for representing the relevance and significance of the Malware/Goodware (MG) datasets, through the use of a pre-trained language model called MalBERTv2. Our model is trained on publicly available datasets, with a focus on the source code of the apps by extracting the top-ranked files that present the most relevant information. These files are then passed through a pre-tokenization feature generator, and the resulting keywords are used to train the tokenizer from scratch. Finally, we apply a classifier usi...
With the widespread use of deep learning in leading systems, it has become the mainstream in the ... more With the widespread use of deep learning in leading systems, it has become the mainstream in the table detection field. Some tables are difficult to detect because of the likely figure layout or the small size. As a solution to the underlined problem, we propose a novel method, called DCTable, to improve Faster R-CNN for table detection. DCTable came up to extract more discriminative features using a backbone with dilated convolutions in order to improve the quality of region proposals. Another main contribution of this paper is the anchors optimization using the Intersection over Union (IoU)-balanced loss to train the RPN and reduce the false positive rate. This is followed by a RoI Align layer, instead of the ROI pooling, to improve the accuracy during mapping table proposal candidates by eliminating the coarse misalignment and introducing the bilinear interpolation in mapping region proposal candidates. Training and testing on a public dataset showed the effectiveness of the algo...
There is currently a huge interest around autonomous vehicles from both industry and academia. Th... more There is currently a huge interest around autonomous vehicles from both industry and academia. This is mainly due to recent advances in machine learning and deep learning, allowing the development of promising methods for autonomous driving. The gap toward full autonomy is incrementally being reduced with essentially three main existing approaches. First, Modular systems that combine a pipeline of methods with each solving one specific sub-task of driving. Second, Direct Perception techniques that directly estimate affordances (car orientation, distances between lane borders, etc) used to compute control commands through a simple logic. Finally, end-to-end frameworks that automatically map raw sensor data to actuation values. The objective of this paper is to review some recent works focusing on end-to-end deep learning models for lane stable driving, as well as some publicly available real world datasets and open-source simulators that enable the development and evaluation of such methods.
2020 IEEE Canadian Conference on Electrical and Computer Engineering (CCECE), 2020
Unmanned aerial vehicles (UAV) are now used for a large number of applications in everyday life. ... more Unmanned aerial vehicles (UAV) are now used for a large number of applications in everyday life. These applications require autonomous navigation which is enabled by the self-localization solution integrated to the UAV. To perform self-localization, most UAVs are relying on a series of sensors combined with a global navigation satellite system (GNSS) in a sensor fusion framework. However, GNSS are using radio signals which are subjected to a large range of outages and interferences. This paper presents a relative visual localization (RVL) approach for GPS-denied environments using a down-facing 2D monocular camera and an inertial measurement unit (IMU). The solution is embedded in an adapted particle filter and use feature points to match images and estimate the localization of the UAV. A new conditional RVL measure is developed in order to leverage spare computation resources available during the data collection when the UAV is still receiving a GNSS signal. An evaluation of six feature point extraction methods is performed using real-world data while varying the number of feature points extracted. The results are promising and the approach has shown to be more efficient and to have fewer limitations than similar approaches in the literature.
This work presents a deep learning framework based on the use of deep convolutional generative ad... more This work presents a deep learning framework based on the use of deep convolutional generative adversarial networks (DCGAN) for infrared face image super-resolution. We use DCGAN for upscaling the images by a factor of 4 × 4, starting at a size of 16 × 16 and obtaining a 64 × 64 face image. Tests are conducted using different infrared face datasets operating in the near-infrared (NIR) and the long-wave infrared (LWIR) spectrum. We can see that the proposed framework performs well and preserves important details of the face. This kind of approach can be very useful in security applications where we can scan faces in the crowd or detect faces at a distance and upscale them for further recognition through an infrared or a multispectral face recognition system.
The article describes a method to stimulate users’ creativity within constraint-based scenarios a... more The article describes a method to stimulate users’ creativity within constraint-based scenarios and OTSM-TRIZ, which allows to define the problems and partial solutions to be solved during the design process in an appropriate manner. The proposed method aims to overcome constraints and problems defined within product development and related organization resources. Indeed, if these constraints are not properly taken into account, the risk of generating unsuccessful and even ineffective solutions can be high. In this work, a method has been defined, based on the OTSM-TRIZ theory: it guides the users toward the problem solution through a mapping of both the problem to solve and the relationships existing among the problems and constraints. A step-by-step approach is used to describe and propose a systematic structure, allowing to link the conceptual solution with specific solution criteria in the automation field. The validation of the proposed method corresponds to a real case study, ...
Proceedings of the 2010 International Conference on Quantitative InfraRed Thermography, 2010
Face recognition in the infrared spectrum has attracted a lot of interest in recent years. Many o... more Face recognition in the infrared spectrum has attracted a lot of interest in recent years. Many of the techniques used in infrared are based on their visible counterpart, especially linear dimensionality reduction techniques like PCA and LDA. In the thermal infrared spectrum, variations can occur between face images of the same individual due to pose, metabolic, time changes, etc. In this work we introduce the use of non linear dimensionality reduction techniques and a probabilistic Bayesian technique for infrared face recognition. These techniques permit to reduce intrapersonal variation, thus making them very interesting for infrared face recognition. A comparative study is conducted in order to evaluate the performance of the proposed techniques for infrared face recognition. Experimental results show that the non linear and probabilistic techniques are promising and lead to interesting results in the infrared spectrum.
Face recognition is an area of computer vision that has attracted a lot of interest from the rese... more Face recognition is an area of computer vision that has attracted a lot of interest from the research community. A growing demand for robust face recognition software in security applications has driven the development of interesting approaches in this field. A large quantity of research ...
Face recognition is an area that has attracted a lot of interest. Much of the research in this fi... more Face recognition is an area that has attracted a lot of interest. Much of the research in this field was conducted using visible images. With visible cameras the recognition is prone to errors due to illumination changes. To avoid the problems encountered in the visible spectrum many authors have proposed the use of infrared. In this paper we give an overview of the state of the art in face recognition using infrared images. Emphasis is given to more recent works. A growing field in this area is multimodal fusion; work conducted in this field is also presented in this paper and publicly available Infrared face image databases are introduced.
In this work we present an efficient approach for physiological features extraction from near inf... more In this work we present an efficient approach for physiological features extraction from near infrared images of the hand and the lower forearm-wrist region. The physiological features represent the dorsal venous network of the hand and the superficial veins in the lower forearm and wrist region. These networks are unique to each individual and can be used as a biometric
Accurate segmentation of the lungs in CXR images is the basis for an automated CXR image analysis... more Accurate segmentation of the lungs in CXR images is the basis for an automated CXR image analysis system. It helps radiologists in detecting lung areas, subtle signs of disease and improving the diagnosis process for patients. However, precise semantic segmentation of lungs is considered a challenging case due to the presence of the edge rib cage, wide variation of lung shape, and lungs affected by diseases. In this paper, we address the problem of lung segmentation in healthy and unhealthy CXR images. Five models were developed and used in detecting and segmenting lung regions. Two loss functions and three benchmark datasets were employed to evaluate these models. Experimental results showed that the proposed models were able to extract salient global and local features from the input CXR images. The best performing model achieved an F1 score of 97.47%, outperforming recent published models. They proved their ability to separate lung regions from the rib cage and clavicle edges and segment varying lung shape depending on age and gender, as well as challenging cases of lungs affected by anomalies such as tuberculosis and the presence of nodules.
The simultaneous advances in deep learning and the Internet of Things (IoT) have benefited distri... more The simultaneous advances in deep learning and the Internet of Things (IoT) have benefited distributed deep learning paradigms. Federated learning is one of the most promising frameworks, where a server works with local learners to train a global model. The intrinsic heterogeneity of IoT devices, or non-independent and identically distributed (Non-I.I.D.) data, combined with the unstable communication network environment, causes a bottleneck that slows convergence and degrades learning efficiency. Additionally, the majority of weight averaging-based model aggregation approaches raise questions about learning fairness. In this paper, we propose a peer-to-peer federated learning (P2PFL) framework based on Vision Transformers (ViT) models to help solve some of the above issues and classify COVID-19 vs. normal cases on Chest-X-Ray (CXR) images. Particularly, clients jointly iterate and aggregate the models in order to build a robust model. The experimental results demonstrate that the p...
X-ray images are the most widely used medical imaging modality. They are affordable, non-dangerou... more X-ray images are the most widely used medical imaging modality. They are affordable, non-dangerous, accessible, and can be used to identify different diseases. Multiple computer-aided detection (CAD) systems using deep learning (DL) algorithms were recently proposed to support radiologists in identifying different diseases on medical images. In this paper, we propose a novel two-step approach for chest disease classification. The first is a multi-class classification step based on classifying X-ray images by infected organs into three classes (normal, lung disease, and heart disease). The second step of our approach is a binary classification of seven specific lungs and heart diseases. We use a consolidated dataset of 26,316 chest X-ray (CXR) images. Two deep learning methods are proposed in this paper. The first is called DC-ChestNet. It is based on ensembling deep convolutional neural network (DCNN) models. The second is named VT-ChestNet. It is based on a modified transformer model. VT-ChestNet achieved the best performance overcoming DC-ChestNet and state-of-the-art models (DenseNet121, DenseNet201, EfficientNetB5, and Xception). VT-ChestNet obtained an area under curve (AUC) of 95.13% for the first step. For the second step, it obtained an average AUC of 99.26% for heart diseases and an average AUC of 99.57% for lung diseases.
Wildland fires are one of the most dangerous natural risks, causing significant economic damage a... more Wildland fires are one of the most dangerous natural risks, causing significant economic damage and loss of lives worldwide. Every year, millions of hectares are lost, and experts warn that the frequency and severity of wildfires will increase in the coming years due to climate change. To mitigate these hazards, numerous deep learning models were developed to detect and map wildland fires, estimate their severity, and predict their spread. In this paper, we provide a comprehensive review of recent deep learning techniques for detecting, mapping, and predicting wildland fires using satellite remote sensing data. We begin by introducing remote sensing satellite systems and their use in wildfire monitoring. Next, we review the deep learning methods employed for these tasks, including fire detection and mapping, severity estimation, and spread prediction. We further present the popular datasets used in these studies. Finally, we address the challenges faced by these models to accurately...
The COVID-19 virus has made a huge impact on people’s lives ever since the outbreak happened in D... more The COVID-19 virus has made a huge impact on people’s lives ever since the outbreak happened in December 2019. Unfortunately, the COVID-19 virus has not completely vanished from the world yet, and thus, global agitation is still increasing with mutations and variants of the same. Early diagnosis is the best way to decline the mortality risk associated with it. This urges the necessity of developing new computational approaches that can analyze a large dataset and predict the disease in time. Currently, automated virus diagnosis is a major area of research for accurate and timely predictions. Artificial intelligent (AI)-based techniques such as machine learning (ML) and deep learning (DL) can be deployed for this purpose. In this, compared to traditional machine learning techniques, deep Learning approaches show prominent results. Yet it still requires optimization in terms of complex space problems. To address this issue, the proposed method combines deep learning predictive models ...
The world has seen an increase in the number of wildland fires in recent years due to various fac... more The world has seen an increase in the number of wildland fires in recent years due to various factors. Experts warn that the number of wildland fires will continue to increase in the coming years, mainly because of climate change. Numerous safety mechanisms such as remote fire detection systems based on deep learning models and vision transformers have been developed recently, showing promising solutions for these tasks. To the best of our knowledge, there are a limited number of published studies in the literature, which address the implementation of deep learning models for wildland fire classification, detection, and segmentation tasks. As such, in this paper, we present an up-to-date and comprehensive review and analysis of these vision methods and their performances. First, previous works related to wildland fire classification, detection, and segmentation based on deep learning including vision transformers are reviewed. Then, the most popular and public datasets used for thes...
To proactively mitigate malware threats, cybersecurity tools, such as anti-virus and anti-malware... more To proactively mitigate malware threats, cybersecurity tools, such as anti-virus and anti-malware software, as well as firewalls, require frequent updates and proactive implementation. However, processing the vast amounts of dataset examples can be overwhelming when relying solely on traditional methods. In cybersecurity workflows, recent advances in natural language processing (NLP) models can aid in proactively detecting various threats. In this paper, we present a novel approach for representing the relevance and significance of the Malware/Goodware (MG) datasets, through the use of a pre-trained language model called MalBERTv2. Our model is trained on publicly available datasets, with a focus on the source code of the apps by extracting the top-ranked files that present the most relevant information. These files are then passed through a pre-tokenization feature generator, and the resulting keywords are used to train the tokenizer from scratch. Finally, we apply a classifier usi...
With the widespread use of deep learning in leading systems, it has become the mainstream in the ... more With the widespread use of deep learning in leading systems, it has become the mainstream in the table detection field. Some tables are difficult to detect because of the likely figure layout or the small size. As a solution to the underlined problem, we propose a novel method, called DCTable, to improve Faster R-CNN for table detection. DCTable came up to extract more discriminative features using a backbone with dilated convolutions in order to improve the quality of region proposals. Another main contribution of this paper is the anchors optimization using the Intersection over Union (IoU)-balanced loss to train the RPN and reduce the false positive rate. This is followed by a RoI Align layer, instead of the ROI pooling, to improve the accuracy during mapping table proposal candidates by eliminating the coarse misalignment and introducing the bilinear interpolation in mapping region proposal candidates. Training and testing on a public dataset showed the effectiveness of the algo...
There is currently a huge interest around autonomous vehicles from both industry and academia. Th... more There is currently a huge interest around autonomous vehicles from both industry and academia. This is mainly due to recent advances in machine learning and deep learning, allowing the development of promising methods for autonomous driving. The gap toward full autonomy is incrementally being reduced with essentially three main existing approaches. First, Modular systems that combine a pipeline of methods with each solving one specific sub-task of driving. Second, Direct Perception techniques that directly estimate affordances (car orientation, distances between lane borders, etc) used to compute control commands through a simple logic. Finally, end-to-end frameworks that automatically map raw sensor data to actuation values. The objective of this paper is to review some recent works focusing on end-to-end deep learning models for lane stable driving, as well as some publicly available real world datasets and open-source simulators that enable the development and evaluation of such methods.
2020 IEEE Canadian Conference on Electrical and Computer Engineering (CCECE), 2020
Unmanned aerial vehicles (UAV) are now used for a large number of applications in everyday life. ... more Unmanned aerial vehicles (UAV) are now used for a large number of applications in everyday life. These applications require autonomous navigation which is enabled by the self-localization solution integrated to the UAV. To perform self-localization, most UAVs are relying on a series of sensors combined with a global navigation satellite system (GNSS) in a sensor fusion framework. However, GNSS are using radio signals which are subjected to a large range of outages and interferences. This paper presents a relative visual localization (RVL) approach for GPS-denied environments using a down-facing 2D monocular camera and an inertial measurement unit (IMU). The solution is embedded in an adapted particle filter and use feature points to match images and estimate the localization of the UAV. A new conditional RVL measure is developed in order to leverage spare computation resources available during the data collection when the UAV is still receiving a GNSS signal. An evaluation of six feature point extraction methods is performed using real-world data while varying the number of feature points extracted. The results are promising and the approach has shown to be more efficient and to have fewer limitations than similar approaches in the literature.
This work presents a deep learning framework based on the use of deep convolutional generative ad... more This work presents a deep learning framework based on the use of deep convolutional generative adversarial networks (DCGAN) for infrared face image super-resolution. We use DCGAN for upscaling the images by a factor of 4 × 4, starting at a size of 16 × 16 and obtaining a 64 × 64 face image. Tests are conducted using different infrared face datasets operating in the near-infrared (NIR) and the long-wave infrared (LWIR) spectrum. We can see that the proposed framework performs well and preserves important details of the face. This kind of approach can be very useful in security applications where we can scan faces in the crowd or detect faces at a distance and upscale them for further recognition through an infrared or a multispectral face recognition system.
The article describes a method to stimulate users’ creativity within constraint-based scenarios a... more The article describes a method to stimulate users’ creativity within constraint-based scenarios and OTSM-TRIZ, which allows to define the problems and partial solutions to be solved during the design process in an appropriate manner. The proposed method aims to overcome constraints and problems defined within product development and related organization resources. Indeed, if these constraints are not properly taken into account, the risk of generating unsuccessful and even ineffective solutions can be high. In this work, a method has been defined, based on the OTSM-TRIZ theory: it guides the users toward the problem solution through a mapping of both the problem to solve and the relationships existing among the problems and constraints. A step-by-step approach is used to describe and propose a systematic structure, allowing to link the conceptual solution with specific solution criteria in the automation field. The validation of the proposed method corresponds to a real case study, ...
Proceedings of the 2010 International Conference on Quantitative InfraRed Thermography, 2010
Face recognition in the infrared spectrum has attracted a lot of interest in recent years. Many o... more Face recognition in the infrared spectrum has attracted a lot of interest in recent years. Many of the techniques used in infrared are based on their visible counterpart, especially linear dimensionality reduction techniques like PCA and LDA. In the thermal infrared spectrum, variations can occur between face images of the same individual due to pose, metabolic, time changes, etc. In this work we introduce the use of non linear dimensionality reduction techniques and a probabilistic Bayesian technique for infrared face recognition. These techniques permit to reduce intrapersonal variation, thus making them very interesting for infrared face recognition. A comparative study is conducted in order to evaluate the performance of the proposed techniques for infrared face recognition. Experimental results show that the non linear and probabilistic techniques are promising and lead to interesting results in the infrared spectrum.
Face recognition is an area of computer vision that has attracted a lot of interest from the rese... more Face recognition is an area of computer vision that has attracted a lot of interest from the research community. A growing demand for robust face recognition software in security applications has driven the development of interesting approaches in this field. A large quantity of research ...
Face recognition is an area that has attracted a lot of interest. Much of the research in this fi... more Face recognition is an area that has attracted a lot of interest. Much of the research in this field was conducted using visible images. With visible cameras the recognition is prone to errors due to illumination changes. To avoid the problems encountered in the visible spectrum many authors have proposed the use of infrared. In this paper we give an overview of the state of the art in face recognition using infrared images. Emphasis is given to more recent works. A growing field in this area is multimodal fusion; work conducted in this field is also presented in this paper and publicly available Infrared face image databases are introduced.
In this work we present an efficient approach for physiological features extraction from near inf... more In this work we present an efficient approach for physiological features extraction from near infrared images of the hand and the lower forearm-wrist region. The physiological features represent the dorsal venous network of the hand and the superficial veins in the lower forearm and wrist region. These networks are unique to each individual and can be used as a biometric
Uploads
Papers by Moulay Akhloufi