Mohammad Shahverdy | Islamic Azad University, Tafresh Campus (original) (raw)

Papers by Mohammad Shahverdy

2022 Sixth International Conference on Smart Cities, Internet of Things and Applications (SCIoT)

Natural language processing is a highly important subcategory in the wide area of artificial inte... more Natural language processing is a highly important subcategory in the wide area of artificial intelligence. Employing appropriate computational algorithms on sophisticated linguistic operations is the aim of natural language processing to extract and create computational theories from languages. In order to achieve this goal, the knowledge of linguists is needed in addition to computer science. In the field of linguistics, the syntactic and semantic relation of words and phrases and the extraction of causation is very significant which the latter is an information retrieval challenge. Recently, there is an increased attention towards the automatic extraction of causation from textual data sets. Although, previous research extracted the casual relations from uninterrupted data sets by using knowledge-based inference technologies and manual coding. Recently, finding comprehensive approaches for detection and extractions of causal arguments is a research area in the field of natural lan...

2021 11th International Conference on Computer Engineering and Knowledge (ICCKE)

Nowadays, the costs of car accidents have a catastrophic impact on human societies. To this end, ... more Nowadays, the costs of car accidents have a catastrophic impact on human societies. To this end, a number of research have been conducted in order to design systems that are able to detect driving violations. Similarly, the present study was conducted to propose an approach that is able to detect and classify the driver’s behavior into two categories, which are labeled as traffic offender and non-offender, using the information collected via vehicle sensors and devices, such as cellphones. Therefore, the velocity signals, acceleration, engine Revolutions Per Minute (RPM), speed limits, and traffic being dispatched from the vehicle, cellphone, OpenStreetMap and google map were utilized to detect the driver's violation. Further, Support Vector Machine (SVM) was used to classify the driver as offender or non-offender, and the violation will be reported to the police via Inter-Vehicular networks.

2021 11th International Conference on Computer Engineering and Knowledge (ICCKE), Oct 28, 2021

Nowadays, the costs of car accidents have a catastrophic impact on human societies. To this end, ... more Nowadays, the costs of car accidents have a catastrophic impact on human societies. To this end, a number of research have been conducted in order to design systems that are able to detect driving violations. Similarly, the present study was conducted to propose an approach that is able to detect and classify the driver’s behavior into two categories, which are labeled as traffic offender and non-offender, using the information collected via vehicle sensors and devices, such as cellphones. Therefore, the velocity signals, acceleration, engine Revolutions Per Minute (RPM), speed limits, and traffic being dispatched from the vehicle, cellphone, OpenStreetMap and google map were utilized to detect the driver's violation. Further, Support Vector Machine (SVM) was used to classify the driver as offender or non-offender, and the violation will be reported to the police via Inter-Vehicular networks.

Advances in Computer Science : an International Journal, 2015

Wireless Mesh network (WMN) is a state of the art networking standard for next generation of wire... more Wireless Mesh network (WMN) is a state of the art networking standard for next generation of wireless network. The construction of these networks is basis of a network of wireless routers witch forwarding each other’s packets in a multi-hop manner. All users in the network can access the internet via Gateways nodes. Because of the high traffic load towards gateway node, it will become congested. A load balancing mechanism is required to balance the traffic among the gateways and prevent the overloading of any gateway. In this paper, weinvestigatedifferent load balancing techniques in wireless mesh networks to avoid congestion in gateways,as well as we survey the effective parameters that is used in these techniques.

Advances in Computer Science : an International Journal, 2015

The development of wireless computer networks and advances in the fabrication of integrated elect... more The development of wireless computer networks and advances in the fabrication of integrated electronic circuits is one of the key elements in making miniature sensors, Makes it possible to use the wireless sensor networks for environmental monitoring in and around the bodies of animals. This precinct of researches is called the wireless research around the body or WBAN and IEEE Institute has assigned two standards to this matter being LEEE.802.15.6 and IEEE.802.15.4. WBAN aim to facilitate, accelerate and improve the accuracy and reliability of medical care and Because of its wide range of challenges, many studies have been devoted to this precinct. according to IEEE.802.15.6 , the topology in WBAN is in star form and one step and two step communications are supported but Due to changes in body position and the different states of the human that body takes (for example walking , running , sitting and …) connecting nodes in one or two step mode via sync or PDA is not always possible ...

Expert Systems with Applications, 2020

Abstract Driver behavior monitoring system as Intelligent Transportation Systems (ITS) have been ... more Abstract Driver behavior monitoring system as Intelligent Transportation Systems (ITS) have been widely exploited to reduce the traffic accidents risk. Most previous methods for monitoring the driver behavior are rely on computer vision techniques. Such methods suffer from violation of privacy and the possibility of spoofing. This paper presents a novel yet efficient deep learning method for analyzing the driver behavior. We have used the driving signals, including acceleration, gravity, throttle, speed, and Revolutions Per Minute (RPM) to recognize five types of driving styles, including normal, aggressive, distracted, drowsy, and drunk driving. To take the advantages of successful deep neural networks on images, we learn a 2D Convolutional Neural Network (CNN) on images constructed from driving signals based on recurrence plot technique. Experimental results confirm that the proposed method can efficiently detect the driver behavior.

2011 IEEE 3rd International Conference on Communication Software and Networks, 2011

Abstract Grid computing presents a new method to distributed computing, which can make a virtual ... more Abstract Grid computing presents a new method to distributed computing, which can make a virtual single image of heterogeneous resources, provide uniform applications and integrate wide spread resources in to transparent resource. Some of students and faculty of Azad university of Tafresh attempt to make a grid infrastructure to achieve above goals and then develope them for grid computing application. That called TafreshGrid. In this paper, we introduce the detail of TafreshGrid project. Our project is devided to three phases: 1) ...

Vehicular Ad hoc Network (VANET)is a kind of mobile Ad Hoc networks (MANETs)that is applied for p... more Vehicular Ad hoc Network (VANET)is a kind of mobile Ad Hoc networks (MANETs)that is applied for progress of intelligent transportation system (ITS).Vehicles get various types of data from roadside units (RSUs)including voice, video and emergency data. Since the number of vehicles as well as the number and size of files are potentially large and with regard to the types of files can be various, QOS-based scheduling is a challenge. We inquire the scheduling algorithm problems of file distribution from RSUs to vehicles in an urban environment and propose a new scheduling scheme which considers the types of data and improved the performance of network. Request of vehicles are inserted in to 4 queues: 1) Data-queue 2) voice-queue 3) video-queue 4) emergency-queue. Some scheduling policies are used to choose one queue and serve the vehicles. Furthermore, we gave high priority to some requests such as video and emergency data to achieve high quality. In this regard, Simulation results indi...

Nowadays, the wireless network technology, as a very cheap alternative Very cheap to create assoc... more Nowadays, the wireless network technology, as a very cheap alternative Very cheap to create associative networks and unified are under development. Although such networks are economically very affordable and are also numerous operational efficiency, including dynamic capability and comfort give and offer to the end user. Among the networks-based infrastructure, a wireless mesh network (as a set of wireless router are taken at strategic locations, providing total connectivity to the network) also provides flexibility for multi-hop. Therefore, how can efficiently these packages through the wireless network arrive to the destination, is a very important problem. Unlike routing wired networks, wireless routing faces with challenges such as interference, among other transfers and channel variable characteristics. In a wireless network, the routing algorithms are divided in numerous categories ranging from algorithms Geographic routing, Geo-Casting, hierarchical, multi-path, based on the ...

Wireless Mesh network (WMN) is a state of the art networking standard for next generation of wire... more Wireless Mesh network (WMN) is a state of the art networking standard for next generation of wireless network. The construction of these networks is basis of a network of wireless routers witch forwarding each other’s packets in a multi-hop manner. All users in the network can access the internet via Gateways nodes. Because of the high traffic load towards gateway node, it will become congested. A load balancing mechanism is required to balance the traffic among the gateways and prevent the overloading of any gateway. In this paper, weinvestigatedifferent load balancing techniques in wireless mesh networks to avoid congestion in gateways,as well as we survey the effective parameters that is used in these techniques.

The development of wireless computer networks and advances in the fabrication of integrated elect... more The development of wireless computer networks and advances in the fabrication of integrated electronic circuits is one of the key elements in making miniature sensors, Makes it possible to use the wireless sensor networks for environmental monitoring in and around the bodies of animals. This precinct of researches is called the wireless research around the body or WBAN and IEEE Institute has assigned two standards to this matter being LEEE.802.15.6 and IEEE.802.15.4. WBAN aim to facilitate, accelerate and improve the accuracy and reliability of medical care and Because of its wide range of challenges, many studies have been devoted to this precinct. according to IEEE.802.15.6 , the topology in WBAN is in star form and one step and two step communications are supported but Due to changes in body position and the different states of the human that body takes (for example walking , running , sitting and …) connecting nodes in one or two step mode via sync or PDA is not always possible ...

In this paper, we address the problem of load balancing in Wireless Mesh Networks. We consider a ... more In this paper, we address the problem of load balancing in Wireless Mesh Networks. We consider a Cluster Based Wireless Mesh Architecture in which the WMN is divided into clusters that could minimize the updating overhead during topology change due to mobility of mesh nodes or congestion of load on a cluster. Each cluster contains a gateway that has complete knowledge about group memberships and link state information in the cluster. The gateway is often elected in the cluster formation process. We consider load of gateways and try to reduce it. As a matter of fact when a gateway undertakes to be an interface for connecting nodes of a wireless mesh network to other networks or internet, there would be some problems such as congestion and bottleneck, so we introduce a new paradigm for these problems. For solving bottleneck we use clustering to reduce load of gateways and after that by use of dividing cluster we prevent from bottleneck on gateways. We study how to detect congestion on a gateway and how can reduce loads of it that preventing from bottleneck on gateway and therefore increasing throughput of network to encountering many loads. So we propose an algorithm to detect bottleneck and remedies for load balancing in Wireless Mesh Networks. We also use Ns2-Emultion for implementing and testing the framework. Some qualitative results are provided to prove the correctness and the advantages of our framework.

... 86 Maryam Asgari Computer Engineering Faculty Islamic Azad University Tafresh branch,lran asg... more ... 86 Maryam Asgari Computer Engineering Faculty Islamic Azad University Tafresh branch,lran asgari@iautb.ac.ir ... NFPSO function algorithm for the evaluation of the solution space to find solutions to existing conditions, there are merit function is called. ...

In mobile ad hoc networks (MANET) due to the high mobility of nodes, routing and load balancing a... more In mobile ad hoc networks (MANET) due to the high mobility of nodes, routing and load balancing are of major problems. In the proposed framework first, routes leading to destination is found and the value of compound route with the average of total probability of nodes for each route is calculated. Final value is considered to select a route among all routes leading to the destination. We considered the parameters in order to balance the load on the network and use of learning automata technique to select the optimal route. Simulation result shows that the proposed approach offers better performance in terms of energy consumption and the number of packets received by the sink.

Binarization or Thresholding is an important basic step in computer vision and image analysis app... more Binarization or Thresholding is an important basic step in computer vision and image analysis applications and has a substantial influence such as, preprocessing step in camera calibration, documents image analysis (OCR) and augmented reality applications. Binarization use thresholding value to separate foreground from the background (convert color or grayscale image to binary image) and reduction the amount of data to be process and raise the computational speed. Recently, there has been an increased interest in multilevel Thresholding. However, as the number of levels increases, the computation time rises. In addition, single threshold methods are faster than multilevel methods because of reduced amount of data to process. Another category of thresholding methods is global and local thresholding. Moreover, for each new application, new methods need to develop due to different requirements of applications. The proposed method is Global Single Fuzzy Threshold based on PSNR called GS...

Binarization or thresholding is one problem that have to solve in pattern recognition methods and... more Binarization or thresholding is one problem that have to solve in pattern recognition methods and applications. Moreover, it has a very important influence on the sequent steps in computer vision applications such as, Optical Character Recognition (OCR), image segmentation, and tracking objects. Ant colony optimization (ACO) is a population-based metaheuristic which use to solve optimizations problems in diverse fields, such as traffic congestion and control, structural optimization, manufacturing, and genomics are presented. In this work, a combination of ant colony, edge detection, and thresholding methods are combine in order to use in OCR system. The algorithm which is used the DIBCO 2009 in printed and a handwritten image was tested. This method has a compare with Kittler and Illingworth's Minimum Error Thresholding, potential difference, max entropy, Pirahansiah’s method and Otsu.

The Journal of Supercomputing

International Journal of Advanced Computer Science, Jun 1, 2014

ABSTRACT

2022 Sixth International Conference on Smart Cities, Internet of Things and Applications (SCIoT)

Natural language processing is a highly important subcategory in the wide area of artificial inte... more Natural language processing is a highly important subcategory in the wide area of artificial intelligence. Employing appropriate computational algorithms on sophisticated linguistic operations is the aim of natural language processing to extract and create computational theories from languages. In order to achieve this goal, the knowledge of linguists is needed in addition to computer science. In the field of linguistics, the syntactic and semantic relation of words and phrases and the extraction of causation is very significant which the latter is an information retrieval challenge. Recently, there is an increased attention towards the automatic extraction of causation from textual data sets. Although, previous research extracted the casual relations from uninterrupted data sets by using knowledge-based inference technologies and manual coding. Recently, finding comprehensive approaches for detection and extractions of causal arguments is a research area in the field of natural lan...

2021 11th International Conference on Computer Engineering and Knowledge (ICCKE)

Nowadays, the costs of car accidents have a catastrophic impact on human societies. To this end, ... more Nowadays, the costs of car accidents have a catastrophic impact on human societies. To this end, a number of research have been conducted in order to design systems that are able to detect driving violations. Similarly, the present study was conducted to propose an approach that is able to detect and classify the driver’s behavior into two categories, which are labeled as traffic offender and non-offender, using the information collected via vehicle sensors and devices, such as cellphones. Therefore, the velocity signals, acceleration, engine Revolutions Per Minute (RPM), speed limits, and traffic being dispatched from the vehicle, cellphone, OpenStreetMap and google map were utilized to detect the driver's violation. Further, Support Vector Machine (SVM) was used to classify the driver as offender or non-offender, and the violation will be reported to the police via Inter-Vehicular networks.

2021 11th International Conference on Computer Engineering and Knowledge (ICCKE), Oct 28, 2021

Nowadays, the costs of car accidents have a catastrophic impact on human societies. To this end, ... more Nowadays, the costs of car accidents have a catastrophic impact on human societies. To this end, a number of research have been conducted in order to design systems that are able to detect driving violations. Similarly, the present study was conducted to propose an approach that is able to detect and classify the driver’s behavior into two categories, which are labeled as traffic offender and non-offender, using the information collected via vehicle sensors and devices, such as cellphones. Therefore, the velocity signals, acceleration, engine Revolutions Per Minute (RPM), speed limits, and traffic being dispatched from the vehicle, cellphone, OpenStreetMap and google map were utilized to detect the driver's violation. Further, Support Vector Machine (SVM) was used to classify the driver as offender or non-offender, and the violation will be reported to the police via Inter-Vehicular networks.

Advances in Computer Science : an International Journal, 2015

Wireless Mesh network (WMN) is a state of the art networking standard for next generation of wire... more Wireless Mesh network (WMN) is a state of the art networking standard for next generation of wireless network. The construction of these networks is basis of a network of wireless routers witch forwarding each other’s packets in a multi-hop manner. All users in the network can access the internet via Gateways nodes. Because of the high traffic load towards gateway node, it will become congested. A load balancing mechanism is required to balance the traffic among the gateways and prevent the overloading of any gateway. In this paper, weinvestigatedifferent load balancing techniques in wireless mesh networks to avoid congestion in gateways,as well as we survey the effective parameters that is used in these techniques.

Advances in Computer Science : an International Journal, 2015

The development of wireless computer networks and advances in the fabrication of integrated elect... more The development of wireless computer networks and advances in the fabrication of integrated electronic circuits is one of the key elements in making miniature sensors, Makes it possible to use the wireless sensor networks for environmental monitoring in and around the bodies of animals. This precinct of researches is called the wireless research around the body or WBAN and IEEE Institute has assigned two standards to this matter being LEEE.802.15.6 and IEEE.802.15.4. WBAN aim to facilitate, accelerate and improve the accuracy and reliability of medical care and Because of its wide range of challenges, many studies have been devoted to this precinct. according to IEEE.802.15.6 , the topology in WBAN is in star form and one step and two step communications are supported but Due to changes in body position and the different states of the human that body takes (for example walking , running , sitting and …) connecting nodes in one or two step mode via sync or PDA is not always possible ...

Expert Systems with Applications, 2020

Abstract Driver behavior monitoring system as Intelligent Transportation Systems (ITS) have been ... more Abstract Driver behavior monitoring system as Intelligent Transportation Systems (ITS) have been widely exploited to reduce the traffic accidents risk. Most previous methods for monitoring the driver behavior are rely on computer vision techniques. Such methods suffer from violation of privacy and the possibility of spoofing. This paper presents a novel yet efficient deep learning method for analyzing the driver behavior. We have used the driving signals, including acceleration, gravity, throttle, speed, and Revolutions Per Minute (RPM) to recognize five types of driving styles, including normal, aggressive, distracted, drowsy, and drunk driving. To take the advantages of successful deep neural networks on images, we learn a 2D Convolutional Neural Network (CNN) on images constructed from driving signals based on recurrence plot technique. Experimental results confirm that the proposed method can efficiently detect the driver behavior.

2011 IEEE 3rd International Conference on Communication Software and Networks, 2011

Abstract Grid computing presents a new method to distributed computing, which can make a virtual ... more Abstract Grid computing presents a new method to distributed computing, which can make a virtual single image of heterogeneous resources, provide uniform applications and integrate wide spread resources in to transparent resource. Some of students and faculty of Azad university of Tafresh attempt to make a grid infrastructure to achieve above goals and then develope them for grid computing application. That called TafreshGrid. In this paper, we introduce the detail of TafreshGrid project. Our project is devided to three phases: 1) ...

Vehicular Ad hoc Network (VANET)is a kind of mobile Ad Hoc networks (MANETs)that is applied for p... more Vehicular Ad hoc Network (VANET)is a kind of mobile Ad Hoc networks (MANETs)that is applied for progress of intelligent transportation system (ITS).Vehicles get various types of data from roadside units (RSUs)including voice, video and emergency data. Since the number of vehicles as well as the number and size of files are potentially large and with regard to the types of files can be various, QOS-based scheduling is a challenge. We inquire the scheduling algorithm problems of file distribution from RSUs to vehicles in an urban environment and propose a new scheduling scheme which considers the types of data and improved the performance of network. Request of vehicles are inserted in to 4 queues: 1) Data-queue 2) voice-queue 3) video-queue 4) emergency-queue. Some scheduling policies are used to choose one queue and serve the vehicles. Furthermore, we gave high priority to some requests such as video and emergency data to achieve high quality. In this regard, Simulation results indi...

Nowadays, the wireless network technology, as a very cheap alternative Very cheap to create assoc... more Nowadays, the wireless network technology, as a very cheap alternative Very cheap to create associative networks and unified are under development. Although such networks are economically very affordable and are also numerous operational efficiency, including dynamic capability and comfort give and offer to the end user. Among the networks-based infrastructure, a wireless mesh network (as a set of wireless router are taken at strategic locations, providing total connectivity to the network) also provides flexibility for multi-hop. Therefore, how can efficiently these packages through the wireless network arrive to the destination, is a very important problem. Unlike routing wired networks, wireless routing faces with challenges such as interference, among other transfers and channel variable characteristics. In a wireless network, the routing algorithms are divided in numerous categories ranging from algorithms Geographic routing, Geo-Casting, hierarchical, multi-path, based on the ...

Wireless Mesh network (WMN) is a state of the art networking standard for next generation of wire... more Wireless Mesh network (WMN) is a state of the art networking standard for next generation of wireless network. The construction of these networks is basis of a network of wireless routers witch forwarding each other’s packets in a multi-hop manner. All users in the network can access the internet via Gateways nodes. Because of the high traffic load towards gateway node, it will become congested. A load balancing mechanism is required to balance the traffic among the gateways and prevent the overloading of any gateway. In this paper, weinvestigatedifferent load balancing techniques in wireless mesh networks to avoid congestion in gateways,as well as we survey the effective parameters that is used in these techniques.

The development of wireless computer networks and advances in the fabrication of integrated elect... more The development of wireless computer networks and advances in the fabrication of integrated electronic circuits is one of the key elements in making miniature sensors, Makes it possible to use the wireless sensor networks for environmental monitoring in and around the bodies of animals. This precinct of researches is called the wireless research around the body or WBAN and IEEE Institute has assigned two standards to this matter being LEEE.802.15.6 and IEEE.802.15.4. WBAN aim to facilitate, accelerate and improve the accuracy and reliability of medical care and Because of its wide range of challenges, many studies have been devoted to this precinct. according to IEEE.802.15.6 , the topology in WBAN is in star form and one step and two step communications are supported but Due to changes in body position and the different states of the human that body takes (for example walking , running , sitting and …) connecting nodes in one or two step mode via sync or PDA is not always possible ...

In this paper, we address the problem of load balancing in Wireless Mesh Networks. We consider a ... more In this paper, we address the problem of load balancing in Wireless Mesh Networks. We consider a Cluster Based Wireless Mesh Architecture in which the WMN is divided into clusters that could minimize the updating overhead during topology change due to mobility of mesh nodes or congestion of load on a cluster. Each cluster contains a gateway that has complete knowledge about group memberships and link state information in the cluster. The gateway is often elected in the cluster formation process. We consider load of gateways and try to reduce it. As a matter of fact when a gateway undertakes to be an interface for connecting nodes of a wireless mesh network to other networks or internet, there would be some problems such as congestion and bottleneck, so we introduce a new paradigm for these problems. For solving bottleneck we use clustering to reduce load of gateways and after that by use of dividing cluster we prevent from bottleneck on gateways. We study how to detect congestion on a gateway and how can reduce loads of it that preventing from bottleneck on gateway and therefore increasing throughput of network to encountering many loads. So we propose an algorithm to detect bottleneck and remedies for load balancing in Wireless Mesh Networks. We also use Ns2-Emultion for implementing and testing the framework. Some qualitative results are provided to prove the correctness and the advantages of our framework.

... 86 Maryam Asgari Computer Engineering Faculty Islamic Azad University Tafresh branch,lran asg... more ... 86 Maryam Asgari Computer Engineering Faculty Islamic Azad University Tafresh branch,lran asgari@iautb.ac.ir ... NFPSO function algorithm for the evaluation of the solution space to find solutions to existing conditions, there are merit function is called. ...

In mobile ad hoc networks (MANET) due to the high mobility of nodes, routing and load balancing a... more In mobile ad hoc networks (MANET) due to the high mobility of nodes, routing and load balancing are of major problems. In the proposed framework first, routes leading to destination is found and the value of compound route with the average of total probability of nodes for each route is calculated. Final value is considered to select a route among all routes leading to the destination. We considered the parameters in order to balance the load on the network and use of learning automata technique to select the optimal route. Simulation result shows that the proposed approach offers better performance in terms of energy consumption and the number of packets received by the sink.

Binarization or Thresholding is an important basic step in computer vision and image analysis app... more Binarization or Thresholding is an important basic step in computer vision and image analysis applications and has a substantial influence such as, preprocessing step in camera calibration, documents image analysis (OCR) and augmented reality applications. Binarization use thresholding value to separate foreground from the background (convert color or grayscale image to binary image) and reduction the amount of data to be process and raise the computational speed. Recently, there has been an increased interest in multilevel Thresholding. However, as the number of levels increases, the computation time rises. In addition, single threshold methods are faster than multilevel methods because of reduced amount of data to process. Another category of thresholding methods is global and local thresholding. Moreover, for each new application, new methods need to develop due to different requirements of applications. The proposed method is Global Single Fuzzy Threshold based on PSNR called GS...

Binarization or thresholding is one problem that have to solve in pattern recognition methods and... more Binarization or thresholding is one problem that have to solve in pattern recognition methods and applications. Moreover, it has a very important influence on the sequent steps in computer vision applications such as, Optical Character Recognition (OCR), image segmentation, and tracking objects. Ant colony optimization (ACO) is a population-based metaheuristic which use to solve optimizations problems in diverse fields, such as traffic congestion and control, structural optimization, manufacturing, and genomics are presented. In this work, a combination of ant colony, edge detection, and thresholding methods are combine in order to use in OCR system. The algorithm which is used the DIBCO 2009 in printed and a handwritten image was tested. This method has a compare with Kittler and Illingworth's Minimum Error Thresholding, potential difference, max entropy, Pirahansiah’s method and Otsu.

The Journal of Supercomputing

International Journal of Advanced Computer Science, Jun 1, 2014

ABSTRACT