Nader Namazi | The Catholic University of America (original) (raw)
Papers by Nader Namazi
Electronic Imaging, 2021
Drowsiness driving is one of the major reasons causing deadly traffic accidents in the United Sta... more Drowsiness driving is one of the major reasons causing deadly traffic accidents in the United States of America. This paper intends to propose a system to detect different levels of drowsiness, which can help drivers to have enough time to handle sleepiness. Furthermore, we use distinct sound alarms to warn the user to prevent early accidents. The basis of the proposed approach is to consider symptoms of drowsiness, including the amount of eye closure, yawning, eye blinking, and head position to classify the level of drowsiness. We design a method to extract eye and mouth features from 68 key points of facial landmark. These features will help the system to detect the level of drowsiness in realtime video stream based on different symptoms. The experiential results show that the average accuracy of the system that has the capability to detect drowsiness intensity scale in different light conditions is approximately 96.6%.
Proceedings of the 33rd Midwest Symposium on Circuits and Systems
ABSTRACT
Proceedings of 36th Midwest Symposium on Circuits and Systems
ABSTRACT
2016 26th International Conference on Field Programmable Logic and Applications (FPL), 2016
2017 8th IEEE Annual Information Technology, Electronics and Mobile Communication Conference (IEMCON), 2017
Efficient utilization of sensor energy to prolong the WSN' lifetime is proposed in this paper... more Efficient utilization of sensor energy to prolong the WSN' lifetime is proposed in this paper. A hybrid PSO Genetic sleep scheduling algorithm for WSNs to escape the local optima trap is the focus of this work. The proposed scheme uses a new parameter: Local Optimum Detector (LOD) for switching from PSO to GA algorithm in order to escape to the local optima trap caused by PSO.
Classifying detections is an important field of study in many disciplines. Typically, data can be... more Classifying detections is an important field of study in many disciplines. Typically, data can be represented in the form of a multidimensional vector defined within some hyperspace (e.g. One may have the sepal length, sepal width, petal length and petal width of an iris flower). One can view many classification problems as processing an unknown data vector in some way that produces an output which correctly categorizes it (e.g. Is the iris flower Iris Setosa, Iris Versicolour or Iris Virginica)? Exemplar based machine learning techniques tackle these problems by learning from representative training data. Several popular algorithms employing these techniques in various ways have been developed and published in the literature. This study explores and develops an innovative exemplar based machine learning methodology which combines clustering techniques with the tools of principal components analysis (PCA) to tackle this problem. Through clustering the methodology segments each class...
2019 2nd World Symposium on Communication Engineering (WSCE)
This work is concerned with the introduction and development of a technique to optimally position... more This work is concerned with the introduction and development of a technique to optimally position a Mobile Sensor (MS) in a location with adequate side lobe Radio Frequency (RF) signal power. The proposed method involves the generation of a database (DB) of side lobe power distribution for different azimuth angles of the downlink transmitted signal. The generated DB is subsequently used to train and test a Machine Learning (ML) multiclass classifier, as well as two distinct Convolution Neural Networks (CNN), to identify the desired MS location. Simulation experiments are performed which indicate a maximum accuracy of 99.25%, 96.56% and 96.10% for 8 different receiver locations.
SPIE Proceedings, 2013
ABSTRACT Free-Space Optical (FSO) communications is a vital area of research due to its important... more ABSTRACT Free-Space Optical (FSO) communications is a vital area of research due to its important advantages of providing a very large bandwidth and relatively low cost of implementation. One of the inherent limitations on the quality of an FSO communication link is the degradation of the received beam due to atmospheric turbulence. This paper is concerned with prototyping a wavelet-based algorithm to remove or reduce the effect of the scintillation noise and other unwanted signal on an FSO link that uses analog frequency modulation. The applicability of these concepts will be demonstrated by providing a real-time prototype using reconfigurable hardware, namely Field Programmable Gate Arrays (FPGA), and high-level design tools such as System Generator for DSP from Xilinx. Our proposed prototype was realized on the Virtex-6 FPGA ML605 board using the XC6VLX240T-1FFG1156 device.
Proceedings., International Conference on Image Processing, 1995
ABSTRACT Image motion estimation has long been under investigation in many fields. Application ex... more ABSTRACT Image motion estimation has long been under investigation in many fields. Application exist in the areas of data compression, filtering, spatiotemporal interpolation and sampling. We develop and present a new iterative scheme, based on the expectation maximization technique, to find the maximum-likelihood estimate of image motion parameters in noisy images. The focus is on image motion coefficient estimation from noisy measurements
[Proceedings] 1992 IEEE International Symposium on Circuits and Systems
The scheme utilizes the steepest ascent routine to search for the maximum likelihood estimate of ... more The scheme utilizes the steepest ascent routine to search for the maximum likelihood estimate of the Karhunen-Loeve coefficients of s(x). 1.
Modeling and Simulation for Defense Systems and Applications X, 2015
This paper utilizes a synchronized Lorenz chaotic drive/response system, which uses Haar filterin... more This paper utilizes a synchronized Lorenz chaotic drive/response system, which uses Haar filtering and appropriate thresholding in order to detect a transmitted random binary message. Using the Lorenz chaotic attractor to obscure the message, the transmission is passed through an Additive White Gaussian (AWG) channel to successfully retrieve the original binary random data. The detection mechanism employs the Haar Wavelet Transform in combating the channel noise. A communication technique using Chaotic Parameter Modulation (CPM) is simulated in Matlab and prototyped on a reconfigurable hardware platform from Xilinx.
This dissertation study focuses on the development of a technique for model dynamic system detect... more This dissertation study focuses on the development of a technique for model dynamic system detection and state estimation of maneuverable targets in radar tracking applications. The new technique incorporates the Expectation and Maximization algorithm with that of the Interacting Multiple Model (IMM) estimator. The main feature of this technique is its ability to estimate the state of a dynamic system with several behavior modes, which can change from one to another. The IMM-EM estimator general structure consists of a bank of Kalman filters for the state cooperating with a filter for the parameters. The IMM-EM estimator is a suboptimal hybrid filter that has an excellent compromise between performance and complexity. Its complexity is nearly linear in the number of models. The IMM-EM algorithm has three major properties: it is recursive, modular, and has fixed computational requirements per cycle. In each cycle, it consists of three major steps: filtering, combination and model det...
Introduction to signal registration. Part 1 Variable time delay estimation: introduction to time ... more Introduction to signal registration. Part 1 Variable time delay estimation: introduction to time delay estimation a new recursive estimator delay estimation based on the MAP criterion delay estimation based on the ML criterion delay estimation based on the MMSE criterion concluding remarks on part one. Part 2 Nonuniform image motion estimation: introduction to image motion estimation motion estimation based on the ML criterion motion estimation based on the MAP criterion motion estimation in transformed domains estimation of the motion coefficients using Kalman filtering concluding remarks on part two. Appendix: an adaptive realization of the generalized maximum likelihood algorithm.
Affine modeled image motion represents a very important class of motion such as reflection, rotat... more Affine modeled image motion represents a very important class of motion such as reflection, rotation, skew, scaling, and transla- tion. The affine motion model is a parametric function, hence, when images undergo this type of motion, the estimation of the motion parameters proves to be efficient and computationally less exhaustive than the pixel by pixel motion estima- tion algorithms. The estimation of the affine motion parameters is accomplished by using the Expectation-Maximization (EM) technique. The (EM) technique employed here converts the maximi- zation of a log-likelihood function from a coupl- ed parameter maximization problem into simpler and decoupled parameter maximization problems. r (x) = s(x) + wl(x) (la) r (x) = s(x-d(x) ) + wz (x) (Ib) where x = (x ,X IT is the spatial vector, d(x) = (dl(x) ,d2 (x) IT is the unknown displacement funct- ion vector, r(x) = (rl(r),rz(x)lT is the observed vector, s(x) is the noise free image, and wl(x) and w (x) are additive white Gauss...
Wireless Sensor Network, 2017
The particularities of Wireless Sensor Networks require specially designed protocols. Nodes in th... more The particularities of Wireless Sensor Networks require specially designed protocols. Nodes in these networks often possess limited access to energy (usually supplied by batteries), which imposes energy constraints. Additionally, WSNs are commonly deployed in monitoring applications, which may intend to cover large areas. Several techniques have been proposed to improve energy-balance, coverage area or both at the same time. In this paper, an alternative solution is presented. It consists of three main components: Fuzzy C-Means for network clustering, a cluster head rotation mechanism and a sleep scheduling algorithm based on a modified version of Particle Swarm Optimization. Results show that this solution is able to provide a configurable routing protocol that offers reduced energy consumption, while keeping highcoverage area.
Electronic Imaging, 2021
Drowsiness driving is one of the major reasons causing deadly traffic accidents in the United Sta... more Drowsiness driving is one of the major reasons causing deadly traffic accidents in the United States of America. This paper intends to propose a system to detect different levels of drowsiness, which can help drivers to have enough time to handle sleepiness. Furthermore, we use distinct sound alarms to warn the user to prevent early accidents. The basis of the proposed approach is to consider symptoms of drowsiness, including the amount of eye closure, yawning, eye blinking, and head position to classify the level of drowsiness. We design a method to extract eye and mouth features from 68 key points of facial landmark. These features will help the system to detect the level of drowsiness in realtime video stream based on different symptoms. The experiential results show that the average accuracy of the system that has the capability to detect drowsiness intensity scale in different light conditions is approximately 96.6%.
Proceedings of the 33rd Midwest Symposium on Circuits and Systems
ABSTRACT
Proceedings of 36th Midwest Symposium on Circuits and Systems
ABSTRACT
2016 26th International Conference on Field Programmable Logic and Applications (FPL), 2016
2017 8th IEEE Annual Information Technology, Electronics and Mobile Communication Conference (IEMCON), 2017
Efficient utilization of sensor energy to prolong the WSN' lifetime is proposed in this paper... more Efficient utilization of sensor energy to prolong the WSN' lifetime is proposed in this paper. A hybrid PSO Genetic sleep scheduling algorithm for WSNs to escape the local optima trap is the focus of this work. The proposed scheme uses a new parameter: Local Optimum Detector (LOD) for switching from PSO to GA algorithm in order to escape to the local optima trap caused by PSO.
Classifying detections is an important field of study in many disciplines. Typically, data can be... more Classifying detections is an important field of study in many disciplines. Typically, data can be represented in the form of a multidimensional vector defined within some hyperspace (e.g. One may have the sepal length, sepal width, petal length and petal width of an iris flower). One can view many classification problems as processing an unknown data vector in some way that produces an output which correctly categorizes it (e.g. Is the iris flower Iris Setosa, Iris Versicolour or Iris Virginica)? Exemplar based machine learning techniques tackle these problems by learning from representative training data. Several popular algorithms employing these techniques in various ways have been developed and published in the literature. This study explores and develops an innovative exemplar based machine learning methodology which combines clustering techniques with the tools of principal components analysis (PCA) to tackle this problem. Through clustering the methodology segments each class...
2019 2nd World Symposium on Communication Engineering (WSCE)
This work is concerned with the introduction and development of a technique to optimally position... more This work is concerned with the introduction and development of a technique to optimally position a Mobile Sensor (MS) in a location with adequate side lobe Radio Frequency (RF) signal power. The proposed method involves the generation of a database (DB) of side lobe power distribution for different azimuth angles of the downlink transmitted signal. The generated DB is subsequently used to train and test a Machine Learning (ML) multiclass classifier, as well as two distinct Convolution Neural Networks (CNN), to identify the desired MS location. Simulation experiments are performed which indicate a maximum accuracy of 99.25%, 96.56% and 96.10% for 8 different receiver locations.
SPIE Proceedings, 2013
ABSTRACT Free-Space Optical (FSO) communications is a vital area of research due to its important... more ABSTRACT Free-Space Optical (FSO) communications is a vital area of research due to its important advantages of providing a very large bandwidth and relatively low cost of implementation. One of the inherent limitations on the quality of an FSO communication link is the degradation of the received beam due to atmospheric turbulence. This paper is concerned with prototyping a wavelet-based algorithm to remove or reduce the effect of the scintillation noise and other unwanted signal on an FSO link that uses analog frequency modulation. The applicability of these concepts will be demonstrated by providing a real-time prototype using reconfigurable hardware, namely Field Programmable Gate Arrays (FPGA), and high-level design tools such as System Generator for DSP from Xilinx. Our proposed prototype was realized on the Virtex-6 FPGA ML605 board using the XC6VLX240T-1FFG1156 device.
Proceedings., International Conference on Image Processing, 1995
ABSTRACT Image motion estimation has long been under investigation in many fields. Application ex... more ABSTRACT Image motion estimation has long been under investigation in many fields. Application exist in the areas of data compression, filtering, spatiotemporal interpolation and sampling. We develop and present a new iterative scheme, based on the expectation maximization technique, to find the maximum-likelihood estimate of image motion parameters in noisy images. The focus is on image motion coefficient estimation from noisy measurements
[Proceedings] 1992 IEEE International Symposium on Circuits and Systems
The scheme utilizes the steepest ascent routine to search for the maximum likelihood estimate of ... more The scheme utilizes the steepest ascent routine to search for the maximum likelihood estimate of the Karhunen-Loeve coefficients of s(x). 1.
Modeling and Simulation for Defense Systems and Applications X, 2015
This paper utilizes a synchronized Lorenz chaotic drive/response system, which uses Haar filterin... more This paper utilizes a synchronized Lorenz chaotic drive/response system, which uses Haar filtering and appropriate thresholding in order to detect a transmitted random binary message. Using the Lorenz chaotic attractor to obscure the message, the transmission is passed through an Additive White Gaussian (AWG) channel to successfully retrieve the original binary random data. The detection mechanism employs the Haar Wavelet Transform in combating the channel noise. A communication technique using Chaotic Parameter Modulation (CPM) is simulated in Matlab and prototyped on a reconfigurable hardware platform from Xilinx.
This dissertation study focuses on the development of a technique for model dynamic system detect... more This dissertation study focuses on the development of a technique for model dynamic system detection and state estimation of maneuverable targets in radar tracking applications. The new technique incorporates the Expectation and Maximization algorithm with that of the Interacting Multiple Model (IMM) estimator. The main feature of this technique is its ability to estimate the state of a dynamic system with several behavior modes, which can change from one to another. The IMM-EM estimator general structure consists of a bank of Kalman filters for the state cooperating with a filter for the parameters. The IMM-EM estimator is a suboptimal hybrid filter that has an excellent compromise between performance and complexity. Its complexity is nearly linear in the number of models. The IMM-EM algorithm has three major properties: it is recursive, modular, and has fixed computational requirements per cycle. In each cycle, it consists of three major steps: filtering, combination and model det...
Introduction to signal registration. Part 1 Variable time delay estimation: introduction to time ... more Introduction to signal registration. Part 1 Variable time delay estimation: introduction to time delay estimation a new recursive estimator delay estimation based on the MAP criterion delay estimation based on the ML criterion delay estimation based on the MMSE criterion concluding remarks on part one. Part 2 Nonuniform image motion estimation: introduction to image motion estimation motion estimation based on the ML criterion motion estimation based on the MAP criterion motion estimation in transformed domains estimation of the motion coefficients using Kalman filtering concluding remarks on part two. Appendix: an adaptive realization of the generalized maximum likelihood algorithm.
Affine modeled image motion represents a very important class of motion such as reflection, rotat... more Affine modeled image motion represents a very important class of motion such as reflection, rotation, skew, scaling, and transla- tion. The affine motion model is a parametric function, hence, when images undergo this type of motion, the estimation of the motion parameters proves to be efficient and computationally less exhaustive than the pixel by pixel motion estima- tion algorithms. The estimation of the affine motion parameters is accomplished by using the Expectation-Maximization (EM) technique. The (EM) technique employed here converts the maximi- zation of a log-likelihood function from a coupl- ed parameter maximization problem into simpler and decoupled parameter maximization problems. r (x) = s(x) + wl(x) (la) r (x) = s(x-d(x) ) + wz (x) (Ib) where x = (x ,X IT is the spatial vector, d(x) = (dl(x) ,d2 (x) IT is the unknown displacement funct- ion vector, r(x) = (rl(r),rz(x)lT is the observed vector, s(x) is the noise free image, and wl(x) and w (x) are additive white Gauss...
Wireless Sensor Network, 2017
The particularities of Wireless Sensor Networks require specially designed protocols. Nodes in th... more The particularities of Wireless Sensor Networks require specially designed protocols. Nodes in these networks often possess limited access to energy (usually supplied by batteries), which imposes energy constraints. Additionally, WSNs are commonly deployed in monitoring applications, which may intend to cover large areas. Several techniques have been proposed to improve energy-balance, coverage area or both at the same time. In this paper, an alternative solution is presented. It consists of three main components: Fuzzy C-Means for network clustering, a cluster head rotation mechanism and a sleep scheduling algorithm based on a modified version of Particle Swarm Optimization. Results show that this solution is able to provide a configurable routing protocol that offers reduced energy consumption, while keeping highcoverage area.