NAJIB BEN AOUN - Academia.edu (original) (raw)
Papers by NAJIB BEN AOUN
A review on kinship verification from facial information
The visual computer/The visual computer, Jun 14, 2024
DenseViT-XGB: A hybrid approach for dates varieties identification
Neurocomputing, May 1, 2024
Computers, materials & continua/Computers, materials & continua (Print), 2024
Face recognition (FR) technology has numerous applications in artificial intelligence including b... more Face recognition (FR) technology has numerous applications in artificial intelligence including biometrics, security, authentication, law enforcement, and surveillance. Deep learning (DL) models, notably convolutional neural networks (CNNs), have shown promising results in the field of FR. However CNNs are easily fooled since they do not encode position and orientation correlations between features. Hinton et al. envisioned Capsule Networks as a more robust design capable of retaining pose information and spatial correlations to recognize objects more like the brain does. Lower-level capsules hold 8-dimensional vectors of attributes like position, hue, texture, and so on, which are routed to higher-level capsules via a new routing by agreement algorithm. This provides capsule networks with viewpoint invariance, which has previously evaded CNNs. This research presents a FR model based on capsule networks that was tested using the LFW dataset, COMSATS face dataset, and own acquired photos using cameras measuring 128 × 128 pixels, 40 × 40 pixels, and 30 × 30 pixels. The trained model outperforms state-ofthe-art algorithms, achieving 95.82% test accuracy and performing well on unseen faces that have been blurred or rotated. Additionally, the suggested model outperformed the recently released approaches on the COMSATS face dataset, achieving a high accuracy of 92.47%. Based on the results of this research as well as previous results, capsule networks perform better than deeper CNNs on unobserved altered data because of their special equivariance properties.
Image super resolution boosting using beta wavelet
Signal, Image and Video Processing, Dec 11, 2023
Ensemble Machine Learning-Based Egg Parasitism Identification for Endangered Bird Conservation
Communications in computer and information science, Dec 31, 2022
Technologies, Jun 20, 2024
Pain assessment has become an important component in modern healthcare systems. It aids medical p... more Pain assessment has become an important component in modern healthcare systems. It aids medical professionals in patient diagnosis and providing the appropriate care and therapy. Conventionally, patients are asked to provide their pain level verbally. However, this subjective method is generally inaccurate, not possible for non-communicative people, can be affected by physiological and environmental factors and is time-consuming, which renders it inefficient in healthcare settings. So, there has been a growing need to build objective, reliable and automatic pain assessment alternatives. In fact, due to the efficiency of facial expressions as pain biomarkers that accurately expand the pain intensity and the power of machine learning methods to effectively learn the subtle nuances of pain expressions and accurately predict pain intensity, automatic pain assessment methods have evolved rapidly. This paper reviews recent spatial facial expressions and machine learning-based pain assessment methods. Moreover, we highlight the pain intensity scales, datasets and method performance evaluation criteria. In addition, these methods' contributions, strengths and limitations will be reported and discussed. Additionally, the review lays the groundwork for further study and improvement for more accurate automatic pain assessment.
Computational Intelligence and Neuroscience, Aug 16, 2022
Recent articles reported a massive increase in frustration among weak students due to the outbrea... more Recent articles reported a massive increase in frustration among weak students due to the outbreak of COVID-19 and Massive Open Online Courses (MOOCs). ese students need to be evaluated to detect possible psychological counseling and extra attention. On the one hand, the literature reports many optimization techniques focusing on existing students' performance prediction systems. On the other hand, psychological works provide insights into massive research ndings focusing on various students' emotions, including frustration. However, the synchronization among these contributions is still a black box, which delays the mathematical modeling of students' frustration. Also, the literature is still limited in using insights of psychology and assumption-based datasets to provide an in-house iterative procedure for modeling students' frustration severity. is paper proposes an optimization technique called the iterative model of frustration severity (IMFS) to explore the black box. It analyzes students' performance via two modules. First, frustration is divided into four outer layers. Second, the students' performance outcome is split into 34 inner layers. e prediction results are iteratively optimized under the umbrella of frustration severity layers through the outer and inner iterations. During validation, the IMFS achieves promising results with various evaluation measures.
IEEE Access
The outbreak of COVID-19 boosted the rapid increase in E-Learning platforms. It also paves the wa... more The outbreak of COVID-19 boosted the rapid increase in E-Learning platforms. It also paves the way for Massive Online Open Courses (MOOCs) to break the record for students' enrollment in online courses. In such circumstances, it is significant to timely identify at-risk students' Cognitive Skills (CS) through an optimized E-Health service. CS is profoundly influenced (negatively and positively) by many human factors, including anxiety and biological age group. Literature has massive findings that correlated CS with anxiety and ageing. However, the earlier studies contributed to CS prediction algorithms are still limited and not up to the mark to efficiently estimate CS under the umbrellas of anxiety and age clusters. The CS prediction system requires an optimization algorithm to mathematical model the influence of age and anxiety clusters. This work predicts students' CS under the influence of anxiety and age clusters, referred to as the Anxiety and Ageing (AA) mathematical model. It solves threefold challenges. First, the study quantizes students' CS, age, and the adverse effects of anxiety. Second, it iteratively computes CS with respect to the anxiety cluster and further revises it under the influence of the age cluster. Third, the study provides a novel data collection method for future researchers by demonstrating assumption-based datasets. The prediction results manifest that the current model achieved excellent precision, recall, and F1 score performance. INDEX TERMS Cognitive skills prediction, data analytics, prediction optimization, data-driven approach, mathematical modeling.
Computational Intelligence and Neuroscience
Recent articles reported a massive increase in frustration among weak students due to the outbrea... more Recent articles reported a massive increase in frustration among weak students due to the outbreak of COVID-19 and Massive Open Online Courses (MOOCs). These students need to be evaluated to detect possible psychological counseling and extra attention. On the one hand, the literature reports many optimization techniques focusing on existing students’ performance prediction systems. On the other hand, psychological works provide insights into massive research findings focusing on various students’ emotions, including frustration. However, the synchronization among these contributions is still a black box, which delays the mathematical modeling of students’ frustration. Also, the literature is still limited in using insights of psychology and assumption-based datasets to provide an in-house iterative procedure for modeling students’ frustration severity. This paper proposes an optimization technique called the iterative model of frustration severity (IMFS) to explore the black box. It...
Advances in Wavelet Theory and Their Applications in Engineering, Physics and Technology, 2012
Improved Very Deep Recurrent Convolutional Neural Network for Object Recognition
Recently, object recognition has been a very active field of interest. The success of the deep le... more Recently, object recognition has been a very active field of interest. The success of the deep learning based methods in recognizing objects has encouraged recent works to follow this approach. In this paper, we propose a very deep recurrent convolutional neural network approach for object recognition. Our approach uses a very deep convolutional neural network reinforced by integrating recurrent connections to the convolutional layers. Besides, the pooling step has been improved by using two main techniques: the Generalizing Pooling and the spatial pyramid pooling. The Generalizing pooling, that replaces the maxpooling layer commonly used in convolutional neural network, combines pooling operations within a hierarchical tree structure. The Spatial Pyramid Pooling which enables the removal of the fixed size constraint of input image has been conducted. In addition, the data augmentation technique has been used to strengthen the training process. Experiments on three object recognition benchmarks dataset: Pascal VOC 2007, CIFAR-10 and CIFAR-100, have shown the success of our approach.
Neurocomputing, Feb 1, 2019
During the last few years, object recognition has received a big of interest in an attempt to mak... more During the last few years, object recognition has received a big of interest in an attempt to make use of the large scale image datasets. Object recognition allows understanding image based on the objects that it contains. The success of the deep learning based methods in recognizing objects has encouraged recent works to follow this approach.
Multimedia Tools and Applications, Apr 9, 2019
In recent years, semantic segmentation has become one of the most active tasks of the computer vi... more In recent years, semantic segmentation has become one of the most active tasks of the computer vision field. Its goal is to group image pixels into semantically meaningful regions. Deep learning methods, in particular those who use convolutional neural network (CNN), have shown a big success for the semantic segmentation task. In this paper, we will introduce a semantic segmentation system using a reinforced fully convolutional densenet with multiscale kernel prediction method. Our main contribution is to build an encoder-decoder based architecture where we increase the width of dense block in the encoder part by conducting recurrent connections inside the dense block. The resulting network structure is called wider dense block where each dense block takes not only the output of the previous layer but also the initial input of the dense block. These recurrent structure emulates the human brain system and helps to strengthen the extraction of the target features. As a result, our network becomes deeper and wider with no additional parameters used because of weights sharing. Moreover, a multiscale convolutional layer has been conducted after the last dense block of the decoder part to perform model averaging over different spatial scales and to provide a more flexible method. This proposed method has been evaluated on two semantic segmentation benchmarks: CamVid and Cityscapes. Our method outperforms many recent works from the state of the art.
IEEE Access
Intraspecific nest parasitism is a phenomenon that attracts the attention of biologists since it ... more Intraspecific nest parasitism is a phenomenon that attracts the attention of biologists since it helps in saving the endangered species such as Slender Billed Gull. The problem comes from the fact that a parasite female lays its eggs in the nest of another female (host) of the same species which causes the abandon of the nest by the host. This behavior causes a significant reduction in future birds number and leads to the expansion of this specie. Thus, there has been an urgent necessity to clean the nest from parasitic eggs. So, our aim is to build an automatic parasitic egg identification system based on egg visual features information. Our system uses deep learning models which have proven their success for image classification. Indeed, our system conduct an egg image's pre-processing phase followed by Fast Beta Wavelet Network (FBWN) to extract the most efficient descriptors (shape, texture, and color). Then, these features will be inputted to the Stacked AutoEncoder for egg classification. Our proposed system, has been evaluated on 91-egg dataset collected from 31 clutches of eggs in Sfax region, Tunisia. Our model has given a parasitic egg identification accuracy of 89.9% which has outperformed the state-of-the-art method and shows the efficiency and the robustness of our system. INDEX TERMS Intraspecific nest parasitism, slender-billed gull, parasitic egg identification, fast beta wavelet network, stacked autoencoder, deep learning.
Journal of Computer Science, 2023
This open-access article is distributed under a Creative Commons Attribution (CC-BY) 4.0 license.
Computational Intelligence and Neuroscience
In this paper, a new classification approach of breast cancer based on Fully Convolutional Networ... more In this paper, a new classification approach of breast cancer based on Fully Convolutional Networks (FCNs) and Beta Wavelet Autoencoder (BWAE) is presented. FCN, as a powerful image segmentation model, is used to extract the relevant information from mammography images. It will identify the relevant zones to model while WAE is used to model the extracted information for these zones. In fact, WAE has proven its superiority to the majority of the features extraction approaches. The fusion of these two techniques have improved the feature extraction phase and this by keeping and modeling only the relevant and useful features for the identification and description of breast masses. The experimental results showed the effectiveness of our proposed method which has given very encouraging results in comparison with the states of the art approaches on the same mammographic image base. A precision rate of 94% for benign and 93% for malignant was achieved with a recall rate of 92% for benign ...
In this paper, we present a scalable and real-time intelligent transportation system based on a b... more In this paper, we present a scalable and real-time intelligent transportation system based on a big data framework. The proposed system allows for the use of existing data from road sensors to better understand traffic flow, traveler behavior, and increase road network performance. Our transportation system is designed to process large-scale stream data to analyze traffic events such as incidents, crashes and congestion. The experiments performed on the public transportation modes of the city of Casablanca in Morocco reveal that the proposed system achieves a significant gain of time, gathers large-scale data from many road sensors and is not expensive in terms of hardware resource consumption.
Journal of WSCG, 2018
In the computer vision field, semantic segmentation represents a very interesting task. Convoluti... more In the computer vision field, semantic segmentation represents a very interesting task. Convolutional Neural Network methods have shown their great performances in comparison with other semantic segmentation methods. In this paper, we propose a multiscale fully convolutional DenseNet approach for semantic segmentation. Our approach is based on the successful fully convolutional DenseNet method. It is reinforced by integrating a multiscale kernel prediction after the last dense block which performs model averaging over different spatial scales and provides more flexibility of our network to presume more information. Experiments on two semantic segmentation benchmarks: CamVid and Cityscapes have shown the effectiveness of our approach which has outperformed many recent works.
Multimedia Tools and Applications, 2019
In recent years, semantic segmentation has become one of the most active tasks of the computer vi... more In recent years, semantic segmentation has become one of the most active tasks of the computer vision field. Its goal is to group image pixels into semantically meaningful regions. Deep learning methods, in particular those who use convolutional neural network (CNN), have shown a big success for the semantic segmentation task. In this paper, we will introduce a semantic segmentation system using a reinforced fully convolutional densenet with multiscale kernel prediction method. Our main contribution is to build an encoder-decoder based architecture where we increase the width of dense block in the encoder part by conducting recurrent connections inside the dense block. The resulting network structure is called wider dense block where each dense block takes not only the output of the previous layer but also the initial input of the dense block. These recurrent structure emulates the human brain system and helps to strengthen the extraction of the target features. As a result, our network becomes deeper and wider with no additional parameters used because of weights sharing. Moreover, a multiscale convolutional layer has been conducted after the last dense block of the decoder part to perform model averaging over different spatial scales and to provide a more flexible method. This proposed method has been evaluated on two semantic segmentation benchmarks: CamVid and Cityscapes. Our method outperforms many recent works from the state of the art.
Bag of frequent subgraphs approach for image classification
Intelligent Data Analysis, 2015
The bag of words approach describes an image as a histogram of visual words. Therefore, the struc... more The bag of words approach describes an image as a histogram of visual words. Therefore, the structural relation between words is lost. Since graphs are well adapted to represent these structural relations, we propose, in this paper, an image classification framework which draws benefit from the efficiency of the graph in modeling structural information and the good classification performances given
A review on kinship verification from facial information
The visual computer/The visual computer, Jun 14, 2024
DenseViT-XGB: A hybrid approach for dates varieties identification
Neurocomputing, May 1, 2024
Computers, materials & continua/Computers, materials & continua (Print), 2024
Face recognition (FR) technology has numerous applications in artificial intelligence including b... more Face recognition (FR) technology has numerous applications in artificial intelligence including biometrics, security, authentication, law enforcement, and surveillance. Deep learning (DL) models, notably convolutional neural networks (CNNs), have shown promising results in the field of FR. However CNNs are easily fooled since they do not encode position and orientation correlations between features. Hinton et al. envisioned Capsule Networks as a more robust design capable of retaining pose information and spatial correlations to recognize objects more like the brain does. Lower-level capsules hold 8-dimensional vectors of attributes like position, hue, texture, and so on, which are routed to higher-level capsules via a new routing by agreement algorithm. This provides capsule networks with viewpoint invariance, which has previously evaded CNNs. This research presents a FR model based on capsule networks that was tested using the LFW dataset, COMSATS face dataset, and own acquired photos using cameras measuring 128 × 128 pixels, 40 × 40 pixels, and 30 × 30 pixels. The trained model outperforms state-ofthe-art algorithms, achieving 95.82% test accuracy and performing well on unseen faces that have been blurred or rotated. Additionally, the suggested model outperformed the recently released approaches on the COMSATS face dataset, achieving a high accuracy of 92.47%. Based on the results of this research as well as previous results, capsule networks perform better than deeper CNNs on unobserved altered data because of their special equivariance properties.
Image super resolution boosting using beta wavelet
Signal, Image and Video Processing, Dec 11, 2023
Ensemble Machine Learning-Based Egg Parasitism Identification for Endangered Bird Conservation
Communications in computer and information science, Dec 31, 2022
Technologies, Jun 20, 2024
Pain assessment has become an important component in modern healthcare systems. It aids medical p... more Pain assessment has become an important component in modern healthcare systems. It aids medical professionals in patient diagnosis and providing the appropriate care and therapy. Conventionally, patients are asked to provide their pain level verbally. However, this subjective method is generally inaccurate, not possible for non-communicative people, can be affected by physiological and environmental factors and is time-consuming, which renders it inefficient in healthcare settings. So, there has been a growing need to build objective, reliable and automatic pain assessment alternatives. In fact, due to the efficiency of facial expressions as pain biomarkers that accurately expand the pain intensity and the power of machine learning methods to effectively learn the subtle nuances of pain expressions and accurately predict pain intensity, automatic pain assessment methods have evolved rapidly. This paper reviews recent spatial facial expressions and machine learning-based pain assessment methods. Moreover, we highlight the pain intensity scales, datasets and method performance evaluation criteria. In addition, these methods' contributions, strengths and limitations will be reported and discussed. Additionally, the review lays the groundwork for further study and improvement for more accurate automatic pain assessment.
Computational Intelligence and Neuroscience, Aug 16, 2022
Recent articles reported a massive increase in frustration among weak students due to the outbrea... more Recent articles reported a massive increase in frustration among weak students due to the outbreak of COVID-19 and Massive Open Online Courses (MOOCs). ese students need to be evaluated to detect possible psychological counseling and extra attention. On the one hand, the literature reports many optimization techniques focusing on existing students' performance prediction systems. On the other hand, psychological works provide insights into massive research ndings focusing on various students' emotions, including frustration. However, the synchronization among these contributions is still a black box, which delays the mathematical modeling of students' frustration. Also, the literature is still limited in using insights of psychology and assumption-based datasets to provide an in-house iterative procedure for modeling students' frustration severity. is paper proposes an optimization technique called the iterative model of frustration severity (IMFS) to explore the black box. It analyzes students' performance via two modules. First, frustration is divided into four outer layers. Second, the students' performance outcome is split into 34 inner layers. e prediction results are iteratively optimized under the umbrella of frustration severity layers through the outer and inner iterations. During validation, the IMFS achieves promising results with various evaluation measures.
IEEE Access
The outbreak of COVID-19 boosted the rapid increase in E-Learning platforms. It also paves the wa... more The outbreak of COVID-19 boosted the rapid increase in E-Learning platforms. It also paves the way for Massive Online Open Courses (MOOCs) to break the record for students' enrollment in online courses. In such circumstances, it is significant to timely identify at-risk students' Cognitive Skills (CS) through an optimized E-Health service. CS is profoundly influenced (negatively and positively) by many human factors, including anxiety and biological age group. Literature has massive findings that correlated CS with anxiety and ageing. However, the earlier studies contributed to CS prediction algorithms are still limited and not up to the mark to efficiently estimate CS under the umbrellas of anxiety and age clusters. The CS prediction system requires an optimization algorithm to mathematical model the influence of age and anxiety clusters. This work predicts students' CS under the influence of anxiety and age clusters, referred to as the Anxiety and Ageing (AA) mathematical model. It solves threefold challenges. First, the study quantizes students' CS, age, and the adverse effects of anxiety. Second, it iteratively computes CS with respect to the anxiety cluster and further revises it under the influence of the age cluster. Third, the study provides a novel data collection method for future researchers by demonstrating assumption-based datasets. The prediction results manifest that the current model achieved excellent precision, recall, and F1 score performance. INDEX TERMS Cognitive skills prediction, data analytics, prediction optimization, data-driven approach, mathematical modeling.
Computational Intelligence and Neuroscience
Recent articles reported a massive increase in frustration among weak students due to the outbrea... more Recent articles reported a massive increase in frustration among weak students due to the outbreak of COVID-19 and Massive Open Online Courses (MOOCs). These students need to be evaluated to detect possible psychological counseling and extra attention. On the one hand, the literature reports many optimization techniques focusing on existing students’ performance prediction systems. On the other hand, psychological works provide insights into massive research findings focusing on various students’ emotions, including frustration. However, the synchronization among these contributions is still a black box, which delays the mathematical modeling of students’ frustration. Also, the literature is still limited in using insights of psychology and assumption-based datasets to provide an in-house iterative procedure for modeling students’ frustration severity. This paper proposes an optimization technique called the iterative model of frustration severity (IMFS) to explore the black box. It...
Advances in Wavelet Theory and Their Applications in Engineering, Physics and Technology, 2012
Improved Very Deep Recurrent Convolutional Neural Network for Object Recognition
Recently, object recognition has been a very active field of interest. The success of the deep le... more Recently, object recognition has been a very active field of interest. The success of the deep learning based methods in recognizing objects has encouraged recent works to follow this approach. In this paper, we propose a very deep recurrent convolutional neural network approach for object recognition. Our approach uses a very deep convolutional neural network reinforced by integrating recurrent connections to the convolutional layers. Besides, the pooling step has been improved by using two main techniques: the Generalizing Pooling and the spatial pyramid pooling. The Generalizing pooling, that replaces the maxpooling layer commonly used in convolutional neural network, combines pooling operations within a hierarchical tree structure. The Spatial Pyramid Pooling which enables the removal of the fixed size constraint of input image has been conducted. In addition, the data augmentation technique has been used to strengthen the training process. Experiments on three object recognition benchmarks dataset: Pascal VOC 2007, CIFAR-10 and CIFAR-100, have shown the success of our approach.
Neurocomputing, Feb 1, 2019
During the last few years, object recognition has received a big of interest in an attempt to mak... more During the last few years, object recognition has received a big of interest in an attempt to make use of the large scale image datasets. Object recognition allows understanding image based on the objects that it contains. The success of the deep learning based methods in recognizing objects has encouraged recent works to follow this approach.
Multimedia Tools and Applications, Apr 9, 2019
In recent years, semantic segmentation has become one of the most active tasks of the computer vi... more In recent years, semantic segmentation has become one of the most active tasks of the computer vision field. Its goal is to group image pixels into semantically meaningful regions. Deep learning methods, in particular those who use convolutional neural network (CNN), have shown a big success for the semantic segmentation task. In this paper, we will introduce a semantic segmentation system using a reinforced fully convolutional densenet with multiscale kernel prediction method. Our main contribution is to build an encoder-decoder based architecture where we increase the width of dense block in the encoder part by conducting recurrent connections inside the dense block. The resulting network structure is called wider dense block where each dense block takes not only the output of the previous layer but also the initial input of the dense block. These recurrent structure emulates the human brain system and helps to strengthen the extraction of the target features. As a result, our network becomes deeper and wider with no additional parameters used because of weights sharing. Moreover, a multiscale convolutional layer has been conducted after the last dense block of the decoder part to perform model averaging over different spatial scales and to provide a more flexible method. This proposed method has been evaluated on two semantic segmentation benchmarks: CamVid and Cityscapes. Our method outperforms many recent works from the state of the art.
IEEE Access
Intraspecific nest parasitism is a phenomenon that attracts the attention of biologists since it ... more Intraspecific nest parasitism is a phenomenon that attracts the attention of biologists since it helps in saving the endangered species such as Slender Billed Gull. The problem comes from the fact that a parasite female lays its eggs in the nest of another female (host) of the same species which causes the abandon of the nest by the host. This behavior causes a significant reduction in future birds number and leads to the expansion of this specie. Thus, there has been an urgent necessity to clean the nest from parasitic eggs. So, our aim is to build an automatic parasitic egg identification system based on egg visual features information. Our system uses deep learning models which have proven their success for image classification. Indeed, our system conduct an egg image's pre-processing phase followed by Fast Beta Wavelet Network (FBWN) to extract the most efficient descriptors (shape, texture, and color). Then, these features will be inputted to the Stacked AutoEncoder for egg classification. Our proposed system, has been evaluated on 91-egg dataset collected from 31 clutches of eggs in Sfax region, Tunisia. Our model has given a parasitic egg identification accuracy of 89.9% which has outperformed the state-of-the-art method and shows the efficiency and the robustness of our system. INDEX TERMS Intraspecific nest parasitism, slender-billed gull, parasitic egg identification, fast beta wavelet network, stacked autoencoder, deep learning.
Journal of Computer Science, 2023
This open-access article is distributed under a Creative Commons Attribution (CC-BY) 4.0 license.
Computational Intelligence and Neuroscience
In this paper, a new classification approach of breast cancer based on Fully Convolutional Networ... more In this paper, a new classification approach of breast cancer based on Fully Convolutional Networks (FCNs) and Beta Wavelet Autoencoder (BWAE) is presented. FCN, as a powerful image segmentation model, is used to extract the relevant information from mammography images. It will identify the relevant zones to model while WAE is used to model the extracted information for these zones. In fact, WAE has proven its superiority to the majority of the features extraction approaches. The fusion of these two techniques have improved the feature extraction phase and this by keeping and modeling only the relevant and useful features for the identification and description of breast masses. The experimental results showed the effectiveness of our proposed method which has given very encouraging results in comparison with the states of the art approaches on the same mammographic image base. A precision rate of 94% for benign and 93% for malignant was achieved with a recall rate of 92% for benign ...
In this paper, we present a scalable and real-time intelligent transportation system based on a b... more In this paper, we present a scalable and real-time intelligent transportation system based on a big data framework. The proposed system allows for the use of existing data from road sensors to better understand traffic flow, traveler behavior, and increase road network performance. Our transportation system is designed to process large-scale stream data to analyze traffic events such as incidents, crashes and congestion. The experiments performed on the public transportation modes of the city of Casablanca in Morocco reveal that the proposed system achieves a significant gain of time, gathers large-scale data from many road sensors and is not expensive in terms of hardware resource consumption.
Journal of WSCG, 2018
In the computer vision field, semantic segmentation represents a very interesting task. Convoluti... more In the computer vision field, semantic segmentation represents a very interesting task. Convolutional Neural Network methods have shown their great performances in comparison with other semantic segmentation methods. In this paper, we propose a multiscale fully convolutional DenseNet approach for semantic segmentation. Our approach is based on the successful fully convolutional DenseNet method. It is reinforced by integrating a multiscale kernel prediction after the last dense block which performs model averaging over different spatial scales and provides more flexibility of our network to presume more information. Experiments on two semantic segmentation benchmarks: CamVid and Cityscapes have shown the effectiveness of our approach which has outperformed many recent works.
Multimedia Tools and Applications, 2019
In recent years, semantic segmentation has become one of the most active tasks of the computer vi... more In recent years, semantic segmentation has become one of the most active tasks of the computer vision field. Its goal is to group image pixels into semantically meaningful regions. Deep learning methods, in particular those who use convolutional neural network (CNN), have shown a big success for the semantic segmentation task. In this paper, we will introduce a semantic segmentation system using a reinforced fully convolutional densenet with multiscale kernel prediction method. Our main contribution is to build an encoder-decoder based architecture where we increase the width of dense block in the encoder part by conducting recurrent connections inside the dense block. The resulting network structure is called wider dense block where each dense block takes not only the output of the previous layer but also the initial input of the dense block. These recurrent structure emulates the human brain system and helps to strengthen the extraction of the target features. As a result, our network becomes deeper and wider with no additional parameters used because of weights sharing. Moreover, a multiscale convolutional layer has been conducted after the last dense block of the decoder part to perform model averaging over different spatial scales and to provide a more flexible method. This proposed method has been evaluated on two semantic segmentation benchmarks: CamVid and Cityscapes. Our method outperforms many recent works from the state of the art.
Bag of frequent subgraphs approach for image classification
Intelligent Data Analysis, 2015
The bag of words approach describes an image as a histogram of visual words. Therefore, the struc... more The bag of words approach describes an image as a histogram of visual words. Therefore, the structural relation between words is lost. Since graphs are well adapted to represent these structural relations, we propose, in this paper, an image classification framework which draws benefit from the efficiency of the graph in modeling structural information and the good classification performances given