An approximate subgradient algorithm for unconstrained nonsmooth, nonconvex optimization
- Bagirov, Adil, Ganjehlou, Asef Nazari
- Authors: Bagirov, Adil , Ganjehlou, Asef Nazari
- Date: 2008
- Type: Text , Journal article
- Relation: Mathematical Methods of Operations Research Vol. 67, no. 2 (2008), p. 187-206
- Relation: http://purl.org/au-research/grants/arc/DP0666061
- Full Text:
- Reviewed:
- Description: In this paper a new algorithm for minimizing locally Lipschitz functions is developed. Descent directions in this algorithm are computed by solving a system of linear inequalities. The convergence of the algorithm is proved for quasidifferentiable semismooth functions. We present the results of numerical experiments with both regular and nonregular objective functions. We also compare the proposed algorithm with two different versions of the subgradient method using the results of numerical experiments. These results demonstrate the superiority of the proposed algorithm over the subgradient method. © 2007 Springer-Verlag.
- Description: C1
- Authors: Bagirov, Adil , Ganjehlou, Asef Nazari
- Date: 2008
- Type: Text , Journal article
- Relation: Mathematical Methods of Operations Research Vol. 67, no. 2 (2008), p. 187-206
- Relation: http://purl.org/au-research/grants/arc/DP0666061
- Full Text:
- Reviewed:
- Description: In this paper a new algorithm for minimizing locally Lipschitz functions is developed. Descent directions in this algorithm are computed by solving a system of linear inequalities. The convergence of the algorithm is proved for quasidifferentiable semismooth functions. We present the results of numerical experiments with both regular and nonregular objective functions. We also compare the proposed algorithm with two different versions of the subgradient method using the results of numerical experiments. These results demonstrate the superiority of the proposed algorithm over the subgradient method. © 2007 Springer-Verlag.
- Description: C1
Reducible polytopes
- Authors: Yost, David
- Date: 2005
- Type: Text , Conference paper
- Relation: Paper presented at the Sixteenth Australasian Workshop on Combinatorial Algorithms, 18-21 September 2005, Ballarat, Australia, Ballarat, Victoria : 18th - 21st September, 2005
- Full Text:
- Reviewed:
- Description: E1
- Description: 2003001436
- Authors: Yost, David
- Date: 2005
- Type: Text , Conference paper
- Relation: Paper presented at the Sixteenth Australasian Workshop on Combinatorial Algorithms, 18-21 September 2005, Ballarat, Australia, Ballarat, Victoria : 18th - 21st September, 2005
- Full Text:
- Reviewed:
- Description: E1
- Description: 2003001436
Local optimization method with global multidimensional search
- Bagirov, Adil, Rubinov, Alex, Zhang, Jiapu
- Authors: Bagirov, Adil , Rubinov, Alex , Zhang, Jiapu
- Date: 2005
- Type: Text , Journal article
- Relation: Journal of Global Optimization Vol. 32, no. 2 (2005), p. 161-179
- Full Text:
- Reviewed:
- Description: This paper presents a new method for solving global optimization problems. We use a local technique based on the notion of discrete gradients for finding a cone of descent directions and then we use a global cutting angle algorithm for finding global minimum within the intersection of the cone and the feasible region. We present results of numerical experiments with well-known test problems and with the so-called cluster function. These results confirm that the proposed algorithms allows one to find a global minimizer or at least a deep local minimizer of a function with a huge amount of shallow local minima. © Springer 2005.
- Description: C1
- Description: 2003001351
- Authors: Bagirov, Adil , Rubinov, Alex , Zhang, Jiapu
- Date: 2005
- Type: Text , Journal article
- Relation: Journal of Global Optimization Vol. 32, no. 2 (2005), p. 161-179
- Full Text:
- Reviewed:
- Description: This paper presents a new method for solving global optimization problems. We use a local technique based on the notion of discrete gradients for finding a cone of descent directions and then we use a global cutting angle algorithm for finding global minimum within the intersection of the cone and the feasible region. We present results of numerical experiments with well-known test problems and with the so-called cluster function. These results confirm that the proposed algorithms allows one to find a global minimizer or at least a deep local minimizer of a function with a huge amount of shallow local minima. © Springer 2005.
- Description: C1
- Description: 2003001351
Nonsmooth optimization models and algorithms for data clustering and visualization
- Authors: Mohebi, Ehsan
- Date: 2015
- Type: Text , Thesis
- Full Text:
- Description: Cluster analysis deals with the problem of organization of a collection of patterns into clusters based on a similarity measure. Various distance functions can be used to define this measure. Clustering problems with the similarity measure defined by the squared Euclidean distance have been studied extensively over the last five decades. However, problems with other Minkowski norms have attracted significantly less attention. The use of different similarity measures may help to identify different cluster structures of a data set. This in turn may help to significantly improve the decision making process. High dimensional data visualization is another important task in the field of data mining and pattern recognition. To date, the principal component analysis and the self-organizing maps techniques have been used to solve such problems. In this thesis we develop algorithms for solving clustering problems in large data sets using various similarity measures. Such similarity measures are based on the squared L
- Description: Doctor of Philosophy
- Authors: Mohebi, Ehsan
- Date: 2015
- Type: Text , Thesis
- Full Text:
- Description: Cluster analysis deals with the problem of organization of a collection of patterns into clusters based on a similarity measure. Various distance functions can be used to define this measure. Clustering problems with the similarity measure defined by the squared Euclidean distance have been studied extensively over the last five decades. However, problems with other Minkowski norms have attracted significantly less attention. The use of different similarity measures may help to identify different cluster structures of a data set. This in turn may help to significantly improve the decision making process. High dimensional data visualization is another important task in the field of data mining and pattern recognition. To date, the principal component analysis and the self-organizing maps techniques have been used to solve such problems. In this thesis we develop algorithms for solving clustering problems in large data sets using various similarity measures. Such similarity measures are based on the squared L
- Description: Doctor of Philosophy
Hyperbolic smoothing in nonsmooth optimization and applications
- Authors: Al Nuaimat, Alia
- Date: 2014
- Type: Text , Thesis , PhD
- Full Text:
- Description: Nonsmooth nonconvex optimization problems arise in many applications including economics, business and data mining. In these applications objective functions are not necessarily differentiable or convex. Many algorithms have been proposed over the past three decades to solve such problems. In spite of the significant growth in this field, the development of efficient algorithms for solving this kind of problem is still a challenging task. The subgradient method is one of the simplest methods developed for solving these problems. Its convergence was proved only for convex objective functions. This method does not involve any subproblems, neither for finding search directions nor for computation of step lengths, which are fixed ahead of time. Bundle methods and their various modifications are among the most efficient methods for solving nonsmooth optimization problems. These methods involve a quadratic programming subproblem to find search directions. The size of the subproblem may increase significantly with the number of variables, which makes the bundle-type methods unsuitable for large scale nonsmooth optimization problems. The implementation of bundle-type methods, which require the use of the quadratic programming solvers, is not as easy as the implementation of the subgradient methods. Therefore it is beneficial to develop algorithms for nonsmooth nonconvex optimization which are easy to implement and more efficient than the subgradient methods. In this thesis, we develop two new algorithms for solving nonsmooth nonconvex optimization problems based on the use of the hyperbolic smoothing technique and apply them to solve the pumping cost minimization problem in water distribution. Both algorithms use smoothing techniques. The first algorithm is designed for solving finite minimax problems. In order to apply the hyperbolic smoothing we reformulate the objective function in the minimax problem and study the relationship between the original minimax and reformulated problems. We also study the main properties of the hyperbolic smoothing function. Based on these results an algorithm for solving the finite minimax problem is proposed and this algorithm is implemented in GAMS. We present preliminary results of numerical experiments with well-known nonsmooth optimization test problems. We also compare the proposed algorithm with the algorithm that uses the exponential smoothing function as well as with the algorithm based on nonlinear programming reformulation of the finite minimax problem. The second nonsmooth optimization algorithm we developed was used to demonstrate how smooth optimization methods can be applied to solve general nonsmooth (nonconvex) optimization problems. In order to do so we compute subgradients from some neighborhood of the current point and define a system of linear inequalities using these subgradients. Search directions are computed by solving this system. This system is solved by reducing it to the minimization of the convex piecewise linear function over the unit ball. Then the hyperbolic smoothing function is applied to approximate this minimization problem by a sequence of smooth problems which are solved by smooth optimization methods. Such an approach allows one to apply powerful smooth optimization algorithms for solving nonsmooth optimization problems and extend smoothing techniques for solving general nonsmooth nonconvex optimization problems. The convergence of the algorithm based on this approach is studied. The proposed algorithm was implemented in Fortran 95. Preliminary results of numerical experiments are reported and the proposed algorithm is compared with an other five nonsmooth optimization algorithms. We also implement the algorithm in GAMS and compare it with GAMS solvers using results of numerical experiments.
- Description: Doctor of Philosophy
- Authors: Al Nuaimat, Alia
- Date: 2014
- Type: Text , Thesis , PhD
- Full Text:
- Description: Nonsmooth nonconvex optimization problems arise in many applications including economics, business and data mining. In these applications objective functions are not necessarily differentiable or convex. Many algorithms have been proposed over the past three decades to solve such problems. In spite of the significant growth in this field, the development of efficient algorithms for solving this kind of problem is still a challenging task. The subgradient method is one of the simplest methods developed for solving these problems. Its convergence was proved only for convex objective functions. This method does not involve any subproblems, neither for finding search directions nor for computation of step lengths, which are fixed ahead of time. Bundle methods and their various modifications are among the most efficient methods for solving nonsmooth optimization problems. These methods involve a quadratic programming subproblem to find search directions. The size of the subproblem may increase significantly with the number of variables, which makes the bundle-type methods unsuitable for large scale nonsmooth optimization problems. The implementation of bundle-type methods, which require the use of the quadratic programming solvers, is not as easy as the implementation of the subgradient methods. Therefore it is beneficial to develop algorithms for nonsmooth nonconvex optimization which are easy to implement and more efficient than the subgradient methods. In this thesis, we develop two new algorithms for solving nonsmooth nonconvex optimization problems based on the use of the hyperbolic smoothing technique and apply them to solve the pumping cost minimization problem in water distribution. Both algorithms use smoothing techniques. The first algorithm is designed for solving finite minimax problems. In order to apply the hyperbolic smoothing we reformulate the objective function in the minimax problem and study the relationship between the original minimax and reformulated problems. We also study the main properties of the hyperbolic smoothing function. Based on these results an algorithm for solving the finite minimax problem is proposed and this algorithm is implemented in GAMS. We present preliminary results of numerical experiments with well-known nonsmooth optimization test problems. We also compare the proposed algorithm with the algorithm that uses the exponential smoothing function as well as with the algorithm based on nonlinear programming reformulation of the finite minimax problem. The second nonsmooth optimization algorithm we developed was used to demonstrate how smooth optimization methods can be applied to solve general nonsmooth (nonconvex) optimization problems. In order to do so we compute subgradients from some neighborhood of the current point and define a system of linear inequalities using these subgradients. Search directions are computed by solving this system. This system is solved by reducing it to the minimization of the convex piecewise linear function over the unit ball. Then the hyperbolic smoothing function is applied to approximate this minimization problem by a sequence of smooth problems which are solved by smooth optimization methods. Such an approach allows one to apply powerful smooth optimization algorithms for solving nonsmooth optimization problems and extend smoothing techniques for solving general nonsmooth nonconvex optimization problems. The convergence of the algorithm based on this approach is studied. The proposed algorithm was implemented in Fortran 95. Preliminary results of numerical experiments are reported and the proposed algorithm is compared with an other five nonsmooth optimization algorithms. We also implement the algorithm in GAMS and compare it with GAMS solvers using results of numerical experiments.
- Description: Doctor of Philosophy
Structure learning of Bayesian Networks using global optimization with applications in data classification
- Taheri, Sona, Mammadov, Musa
- Authors: Taheri, Sona , Mammadov, Musa
- Date: 2014
- Type: Text , Journal article
- Relation: Optimization Letters Vol. 9, no. 5 (2014), p. 931-948
- Full Text:
- Reviewed:
- Description: Bayesian Networks are increasingly popular methods of modeling uncertainty in artificial intelligence and machine learning. A Bayesian Network consists of a directed acyclic graph in which each node represents a variable and each arc represents probabilistic dependency between two variables. Constructing a Bayesian Network from data is a learning process that consists of two steps: learning structure and learning parameter. Learning a network structure from data is the most difficult task in this process. This paper presents a new algorithm for constructing an optimal structure for Bayesian Networks based on optimization. The algorithm has two major parts. First, we define an optimization model to find the better network graphs. Then, we apply an optimization approach for removing possible cycles from the directed graphs obtained in the first part which is the first of its kind in the literature. The main advantage of the proposed method is that the maximal number of parents for variables is not fixed a priory and it is defined during the optimization procedure. It also considers all networks including cyclic ones and then choose a best structure by applying a global optimization method. To show the efficiency of the algorithm, several closely related algorithms including unrestricted dependency Bayesian Network algorithm, as well as, benchmarks algorithms SVM and C4.5 are employed for comparison. We apply these algorithms on data classification; data sets are taken from the UCI machine learning repository and the LIBSVM. © 2014, Springer-Verlag Berlin Heidelberg.
- Authors: Taheri, Sona , Mammadov, Musa
- Date: 2014
- Type: Text , Journal article
- Relation: Optimization Letters Vol. 9, no. 5 (2014), p. 931-948
- Full Text:
- Reviewed:
- Description: Bayesian Networks are increasingly popular methods of modeling uncertainty in artificial intelligence and machine learning. A Bayesian Network consists of a directed acyclic graph in which each node represents a variable and each arc represents probabilistic dependency between two variables. Constructing a Bayesian Network from data is a learning process that consists of two steps: learning structure and learning parameter. Learning a network structure from data is the most difficult task in this process. This paper presents a new algorithm for constructing an optimal structure for Bayesian Networks based on optimization. The algorithm has two major parts. First, we define an optimization model to find the better network graphs. Then, we apply an optimization approach for removing possible cycles from the directed graphs obtained in the first part which is the first of its kind in the literature. The main advantage of the proposed method is that the maximal number of parents for variables is not fixed a priory and it is defined during the optimization procedure. It also considers all networks including cyclic ones and then choose a best structure by applying a global optimization method. To show the efficiency of the algorithm, several closely related algorithms including unrestricted dependency Bayesian Network algorithm, as well as, benchmarks algorithms SVM and C4.5 are employed for comparison. We apply these algorithms on data classification; data sets are taken from the UCI machine learning repository and the LIBSVM. © 2014, Springer-Verlag Berlin Heidelberg.
Nonsmooth and derivative-free optimization based hybrid methods and applications
- Authors: Long, Qiang
- Date: 2014
- Type: Text , Thesis , PhD
- Full Text:
- Description: "In this thesis, we develop hybrid methods for solving global and in particular, nonsmooth optimization problems. Hybrid methods are becoming more popular in global optimization since they allow to apply powerful smooth optimization techniques to solve global optimization problems. Such methods are able to efficiently solve global optimization problems with large number of variables. To date global search algorithms have been mainly applied to improve global search properties of the local search methods (including smooth optimization algorithms). In this thesis we apply rather different strategy to design hybrid methods. We use local search algorithms to improve the efficiency of global search methods. The thesis consists of two parts. In the first part we describe hybrid algorithms and in the second part we consider their various applications." -- taken from Abstract.
- Description: Operational Research and Cybernetics
- Authors: Long, Qiang
- Date: 2014
- Type: Text , Thesis , PhD
- Full Text:
- Description: "In this thesis, we develop hybrid methods for solving global and in particular, nonsmooth optimization problems. Hybrid methods are becoming more popular in global optimization since they allow to apply powerful smooth optimization techniques to solve global optimization problems. Such methods are able to efficiently solve global optimization problems with large number of variables. To date global search algorithms have been mainly applied to improve global search properties of the local search methods (including smooth optimization algorithms). In this thesis we apply rather different strategy to design hybrid methods. We use local search algorithms to improve the efficiency of global search methods. The thesis consists of two parts. In the first part we describe hybrid algorithms and in the second part we consider their various applications." -- taken from Abstract.
- Description: Operational Research and Cybernetics
Bio-inspired network security for 5G-enabled IoT applications
- Saleem, Kashif, Alabduljabbar, Ghadah, Alrowais, Nouf, Al-Muhtadi, Jalal, Imran, Muhammad, Rodrigues, Joel
- Authors: Saleem, Kashif , Alabduljabbar, Ghadah , Alrowais, Nouf , Al-Muhtadi, Jalal , Imran, Muhammad , Rodrigues, Joel
- Date: 2020
- Type: Text , Journal article
- Relation: IEEE access Vol. 8, no. (2020), p. 1-1
- Full Text:
- Reviewed:
- Description: Every IPv6-enabled device connected and communicating over the Internet forms the Internet of things (IoT) that is prevalent in society and is used in daily life. This IoT platform will quickly grow to be populated with billions or more objects by making every electrical appliance, car, and even items of furniture smart and connected. The 5th generation (5G) and beyond networks will further boost these IoT systems. The massive utilization of these systems over gigabits per second generates numerous issues. Owing to the huge complexity in large-scale deployment of IoT, data privacy and security are the most prominent challenges, especially for critical applications such as Industry 4.0, e-healthcare, and military. Threat agents persistently strive to find new vulnerabilities and exploit them. Therefore, including promising security measures to support the running systems, not to harm or collapse them, is essential. Nature-inspired algorithms have the capability to provide autonomous and sustainable defense and healing mechanisms. This paper first surveys the 5G network layer security for IoT applications and lists the network layer security vulnerabilities and requirements in wireless sensor networks, IoT, and 5G-enabled IoT. Second, a detailed literature review is conducted with the current network layer security methods and the bio-inspired techniques for IoT applications exchanging data packets over 5G. Finally, the bio-inspired algorithms are analyzed in the context of providing a secure network layer for IoT applications connected over 5G and beyond networks.
- Authors: Saleem, Kashif , Alabduljabbar, Ghadah , Alrowais, Nouf , Al-Muhtadi, Jalal , Imran, Muhammad , Rodrigues, Joel
- Date: 2020
- Type: Text , Journal article
- Relation: IEEE access Vol. 8, no. (2020), p. 1-1
- Full Text:
- Reviewed:
- Description: Every IPv6-enabled device connected and communicating over the Internet forms the Internet of things (IoT) that is prevalent in society and is used in daily life. This IoT platform will quickly grow to be populated with billions or more objects by making every electrical appliance, car, and even items of furniture smart and connected. The 5th generation (5G) and beyond networks will further boost these IoT systems. The massive utilization of these systems over gigabits per second generates numerous issues. Owing to the huge complexity in large-scale deployment of IoT, data privacy and security are the most prominent challenges, especially for critical applications such as Industry 4.0, e-healthcare, and military. Threat agents persistently strive to find new vulnerabilities and exploit them. Therefore, including promising security measures to support the running systems, not to harm or collapse them, is essential. Nature-inspired algorithms have the capability to provide autonomous and sustainable defense and healing mechanisms. This paper first surveys the 5G network layer security for IoT applications and lists the network layer security vulnerabilities and requirements in wireless sensor networks, IoT, and 5G-enabled IoT. Second, a detailed literature review is conducted with the current network layer security methods and the bio-inspired techniques for IoT applications exchanging data packets over 5G. Finally, the bio-inspired algorithms are analyzed in the context of providing a secure network layer for IoT applications connected over 5G and beyond networks.
Algorithm development for the non-destructive testing of structural damage
- Noori Hoshyar, Azadeh, Rashidi, Maria, Liyanapathirana, Ranjith, Samali, Bijan
- Authors: Noori Hoshyar, Azadeh , Rashidi, Maria , Liyanapathirana, Ranjith , Samali, Bijan
- Date: 2019
- Type: Text , Journal article
- Relation: Applied sciences Vol. 9, no. 14 (2019), p. 2810
- Full Text:
- Reviewed:
- Description: Monitoring of structures to identify types of damages that occur under loading is essential in practical applications of civil infrastructure. In this paper, we detect and visualize damage based on several non-destructive testing (NDT) methods. A machine learning (ML) approach based on the Support Vector Machine (SVM) method is developed to prevent misdirection of the event interpretation of what is happening in the material. The objective is to identify cracks in the early stages, to reduce the risk of failure in structures. Theoretical and experimental analyses are derived by computing the performance indicators on the smart aggregate (SA)-based sensor data for concrete and reinforced-concrete (RC) beams. Validity assessment of the proposed indices was addressed through a comparative analysis with traditional SVM. The developed ML algorithms are shown to recognize cracks with a higher accuracy than the traditional SVM. Additionally, we propose different algorithms for microwave- or millimeter-wave imaging of steel plates, composite materials, and metal plates, to identify and visualize cracks. The proposed algorithm for steel plates is based on the gradient magnitude in four directions of an image, and is followed by the edge detection technique. Three algorithms were proposed for each of composite materials and metal plates, and are based on 2D fast Fourier transform (FFT) and hybrid fuzzy c-mean techniques, respectively. The proposed algorithms were able to recognize and visualize the cracking incurred in the structure more efficiently than the traditional techniques. The reported results are expected to be beneficial for NDT-based applications, particularly in civil engineering.
- Authors: Noori Hoshyar, Azadeh , Rashidi, Maria , Liyanapathirana, Ranjith , Samali, Bijan
- Date: 2019
- Type: Text , Journal article
- Relation: Applied sciences Vol. 9, no. 14 (2019), p. 2810
- Full Text:
- Reviewed:
- Description: Monitoring of structures to identify types of damages that occur under loading is essential in practical applications of civil infrastructure. In this paper, we detect and visualize damage based on several non-destructive testing (NDT) methods. A machine learning (ML) approach based on the Support Vector Machine (SVM) method is developed to prevent misdirection of the event interpretation of what is happening in the material. The objective is to identify cracks in the early stages, to reduce the risk of failure in structures. Theoretical and experimental analyses are derived by computing the performance indicators on the smart aggregate (SA)-based sensor data for concrete and reinforced-concrete (RC) beams. Validity assessment of the proposed indices was addressed through a comparative analysis with traditional SVM. The developed ML algorithms are shown to recognize cracks with a higher accuracy than the traditional SVM. Additionally, we propose different algorithms for microwave- or millimeter-wave imaging of steel plates, composite materials, and metal plates, to identify and visualize cracks. The proposed algorithm for steel plates is based on the gradient magnitude in four directions of an image, and is followed by the edge detection technique. Three algorithms were proposed for each of composite materials and metal plates, and are based on 2D fast Fourier transform (FFT) and hybrid fuzzy c-mean techniques, respectively. The proposed algorithms were able to recognize and visualize the cracking incurred in the structure more efficiently than the traditional techniques. The reported results are expected to be beneficial for NDT-based applications, particularly in civil engineering.
Improving Naive Bayes classifier using conditional probabilities
- Taheri, Sona, Mammadov, Musa, Bagirov, Adil
- Authors: Taheri, Sona , Mammadov, Musa , Bagirov, Adil
- Date: 2010
- Type: Text , Conference proceedings
- Full Text:
- Description: Naive Bayes classifier is the simplest among Bayesian Network classifiers. It has shown to be very efficient on a variety of data classification problems. However, the strong assumption that all features are conditionally independent given the class is often violated on many real world applications. Therefore, improvement of the Naive Bayes classifier by alleviating the feature independence assumption has attracted much attention. In this paper, we develop a new version of the Naive Bayes classifier without assuming independence of features. The proposed algorithm approximates the interactions between features by using conditional probabilities. We present results of numerical experiments on several real world data sets, where continuous features are discretized by applying two different methods. These results demonstrate that the proposed algorithm significantly improve the performance of the Naive Bayes classifier, yet at the same time maintains its robustness. © 2011, Australian Computer Society, Inc.
- Description: 2003009505
- Authors: Taheri, Sona , Mammadov, Musa , Bagirov, Adil
- Date: 2010
- Type: Text , Conference proceedings
- Full Text:
- Description: Naive Bayes classifier is the simplest among Bayesian Network classifiers. It has shown to be very efficient on a variety of data classification problems. However, the strong assumption that all features are conditionally independent given the class is often violated on many real world applications. Therefore, improvement of the Naive Bayes classifier by alleviating the feature independence assumption has attracted much attention. In this paper, we develop a new version of the Naive Bayes classifier without assuming independence of features. The proposed algorithm approximates the interactions between features by using conditional probabilities. We present results of numerical experiments on several real world data sets, where continuous features are discretized by applying two different methods. These results demonstrate that the proposed algorithm significantly improve the performance of the Naive Bayes classifier, yet at the same time maintains its robustness. © 2011, Australian Computer Society, Inc.
- Description: 2003009505
A feature agnostic approach for glaucoma detection in OCT volumes
- Maetschke, Stefan, Antony, Bhavna, Ishikawa, Hiroshi, Wollstein, Gadi, Schuman, Joel, Garnavi, Rahil
- Authors: Maetschke, Stefan , Antony, Bhavna , Ishikawa, Hiroshi , Wollstein, Gadi , Schuman, Joel , Garnavi, Rahil
- Date: 2019
- Type: Text , Journal article
- Relation: PLoS One Vol. 14, no. 7 (2019), p. e0219126
- Full Text:
- Reviewed:
- Description: Optical coherence tomography (OCT) based measurements of retinal layer thickness, such as the retinal nerve fibre layer (RNFL) and the ganglion cell with inner plexiform layer (GCIPL) are commonly employed for the diagnosis and monitoring of glaucoma. Previously, machine learning techniques have relied on segmentation-based imaging features such as the peripapillary RNFL thickness and the cup-to-disc ratio. Here, we propose a deep learning technique that classifies eyes as healthy or glaucomatous directly from raw, unsegmented OCT volumes of the optic nerve head (ONH) using a 3D Convolutional Neural Network (CNN). We compared the accuracy of this technique with various feature-based machine learning algorithms and demonstrated the superiority of the proposed deep learning based method. Logistic regression was found to be the best performing classical machine learning technique with an AUC of 0.89. In direct comparison, the deep learning approach achieved a substantially higher AUC of 0.94 with the additional advantage of providing insight into which regions of an OCT volume are important for glaucoma detection. Computing Class Activation Maps (CAM), we found that the CNN identified neuroretinal rim and optic disc cupping as well as the lamina cribrosa (LC) and its surrounding areas as the regions significantly associated with the glaucoma classification. These regions anatomically correspond to the well established and commonly used clinical markers for glaucoma diagnosis such as increased cup volume, cup diameter, and neuroretinal rim thinning at the superior and inferior segments.
- Authors: Maetschke, Stefan , Antony, Bhavna , Ishikawa, Hiroshi , Wollstein, Gadi , Schuman, Joel , Garnavi, Rahil
- Date: 2019
- Type: Text , Journal article
- Relation: PLoS One Vol. 14, no. 7 (2019), p. e0219126
- Full Text:
- Reviewed:
- Description: Optical coherence tomography (OCT) based measurements of retinal layer thickness, such as the retinal nerve fibre layer (RNFL) and the ganglion cell with inner plexiform layer (GCIPL) are commonly employed for the diagnosis and monitoring of glaucoma. Previously, machine learning techniques have relied on segmentation-based imaging features such as the peripapillary RNFL thickness and the cup-to-disc ratio. Here, we propose a deep learning technique that classifies eyes as healthy or glaucomatous directly from raw, unsegmented OCT volumes of the optic nerve head (ONH) using a 3D Convolutional Neural Network (CNN). We compared the accuracy of this technique with various feature-based machine learning algorithms and demonstrated the superiority of the proposed deep learning based method. Logistic regression was found to be the best performing classical machine learning technique with an AUC of 0.89. In direct comparison, the deep learning approach achieved a substantially higher AUC of 0.94 with the additional advantage of providing insight into which regions of an OCT volume are important for glaucoma detection. Computing Class Activation Maps (CAM), we found that the CNN identified neuroretinal rim and optic disc cupping as well as the lamina cribrosa (LC) and its surrounding areas as the regions significantly associated with the glaucoma classification. These regions anatomically correspond to the well established and commonly used clinical markers for glaucoma diagnosis such as increased cup volume, cup diameter, and neuroretinal rim thinning at the superior and inferior segments.
Automated segmentation of mouse OCT volumes (ASiMOV): Validation & clinical study of a light damage model
- Antony, Bhavna, Kim, Byung-Jin, Lang, Andrew, Carass, Aaron, Prince, Jerry, Zack, Donald
- Authors: Antony, Bhavna , Kim, Byung-Jin , Lang, Andrew , Carass, Aaron , Prince, Jerry , Zack, Donald
- Date: 2017
- Type: Text , Journal article
- Relation: PLoS One Vol. 12, no. 8 (2017), p. e0181059-e0181059
- Full Text:
- Reviewed:
- Description: The use of spectral-domain optical coherence tomography (SD-OCT) is becoming commonplace for the in vivo longitudinal study of murine models of ophthalmic disease. Longitudinal studies, however, generate large quantities of data, the manual analysis of which is very challenging due to the time-consuming nature of generating delineations. Thus, it is of importance that automated algorithms be developed to facilitate accurate and timely analysis of these large datasets. Furthermore, as the models target a variety of diseases, the associated structural changes can also be extremely disparate. For instance, in the light damage (LD) model, which is frequently used to study photoreceptor degeneration, the outer retina appears dramatically different from the normal retina. To address these concerns, we have developed a flexible graph-based algorithm for the automated segmentation of mouse OCT volumes (ASiMOV). This approach incorporates a machine-learning component that can be easily trained for different disease models. To validate ASiMOV, the automated results were compared to manual delineations obtained from three raters on healthy and BALB/cJ mice post LD. It was also used to study a longitudinal LD model, where five control and five LD mice were imaged at four timepoints post LD. The total retinal thickness and the outer retina (comprising the outer nuclear layer, and inner and outer segments of the photoreceptors) were unchanged the day after the LD, but subsequently thinned significantly (p < 0.01). The retinal nerve fiber-ganglion cell complex and the inner plexiform layers, however, remained unchanged for the duration of the study.
- Authors: Antony, Bhavna , Kim, Byung-Jin , Lang, Andrew , Carass, Aaron , Prince, Jerry , Zack, Donald
- Date: 2017
- Type: Text , Journal article
- Relation: PLoS One Vol. 12, no. 8 (2017), p. e0181059-e0181059
- Full Text:
- Reviewed:
- Description: The use of spectral-domain optical coherence tomography (SD-OCT) is becoming commonplace for the in vivo longitudinal study of murine models of ophthalmic disease. Longitudinal studies, however, generate large quantities of data, the manual analysis of which is very challenging due to the time-consuming nature of generating delineations. Thus, it is of importance that automated algorithms be developed to facilitate accurate and timely analysis of these large datasets. Furthermore, as the models target a variety of diseases, the associated structural changes can also be extremely disparate. For instance, in the light damage (LD) model, which is frequently used to study photoreceptor degeneration, the outer retina appears dramatically different from the normal retina. To address these concerns, we have developed a flexible graph-based algorithm for the automated segmentation of mouse OCT volumes (ASiMOV). This approach incorporates a machine-learning component that can be easily trained for different disease models. To validate ASiMOV, the automated results were compared to manual delineations obtained from three raters on healthy and BALB/cJ mice post LD. It was also used to study a longitudinal LD model, where five control and five LD mice were imaged at four timepoints post LD. The total retinal thickness and the outer retina (comprising the outer nuclear layer, and inner and outer segments of the photoreceptors) were unchanged the day after the LD, but subsequently thinned significantly (p < 0.01). The retinal nerve fiber-ganglion cell complex and the inner plexiform layers, however, remained unchanged for the duration of the study.
Machine learning-based agoraphilic navigation algorithm for use in dynamic environments with a moving goal
- Hewawasam, Hasitha, Kahandawa, Gayan, Ibrahim, Yousef
- Authors: Hewawasam, Hasitha , Kahandawa, Gayan , Ibrahim, Yousef
- Date: 2023
- Type: Text , Journal article
- Relation: Machines Vol. 11, no. 5 (2023), p. 513
- Full Text:
- Reviewed:
- Description: This paper presents a novel development of a new machine learning-based control system for the Agoraphilic (free-space attraction) concept of navigating robots in unknown dynamic environments with a moving goal. Furthermore, this paper presents a new methodology to generate training and testing datasets to develop a machine learning-based module to improve the performances of Agoraphilic algorithms. The new algorithm presented in this paper utilises the free-space attraction (Agoraphilic) concept to safely navigate a mobile robot in a dynamically cluttered environment with a moving goal. The algorithm uses tracking and prediction strategies to estimate the position and velocity vectors of detected moving obstacles and the goal. This predictive methodology enables the algorithm to identify and incorporate potential future growing free-space passages towards the moving goal. This is supported by the new machine learning-based controller designed specifically to efficiently account for the high uncertainties inherent in the robot’s operational environment with a moving goal at a reduced computational cost. This paper also includes comparative and experimental results to demonstrate the improvements of the algorithm after introducing the machine learning technique. The presented experiments demonstrated the success of the algorithm in navigating robots in dynamic environments with the challenge of a moving goal.
- Authors: Hewawasam, Hasitha , Kahandawa, Gayan , Ibrahim, Yousef
- Date: 2023
- Type: Text , Journal article
- Relation: Machines Vol. 11, no. 5 (2023), p. 513
- Full Text:
- Reviewed:
- Description: This paper presents a novel development of a new machine learning-based control system for the Agoraphilic (free-space attraction) concept of navigating robots in unknown dynamic environments with a moving goal. Furthermore, this paper presents a new methodology to generate training and testing datasets to develop a machine learning-based module to improve the performances of Agoraphilic algorithms. The new algorithm presented in this paper utilises the free-space attraction (Agoraphilic) concept to safely navigate a mobile robot in a dynamically cluttered environment with a moving goal. The algorithm uses tracking and prediction strategies to estimate the position and velocity vectors of detected moving obstacles and the goal. This predictive methodology enables the algorithm to identify and incorporate potential future growing free-space passages towards the moving goal. This is supported by the new machine learning-based controller designed specifically to efficiently account for the high uncertainties inherent in the robot’s operational environment with a moving goal at a reduced computational cost. This paper also includes comparative and experimental results to demonstrate the improvements of the algorithm after introducing the machine learning technique. The presented experiments demonstrated the success of the algorithm in navigating robots in dynamic environments with the challenge of a moving goal.
Comparative analysis of machine and deep learning models for soil properties prediction from hyperspectral visual band
- Datta, Dristi, Paul, Manoranjan, Murshed, Manzur, Teng, Shyh Wei, Schmidtke, Leigh
- Authors: Datta, Dristi , Paul, Manoranjan , Murshed, Manzur , Teng, Shyh Wei , Schmidtke, Leigh
- Date: 2023
- Type: Text , Journal article
- Relation: Environments Vol. 10, no. 5 (2023), p. 77
- Full Text:
- Reviewed:
- Description: Estimating various properties of soil, including moisture, carbon, and nitrogen, is crucial for studying their correlation with plant health and food production. However, conventional methods such as oven-drying and chemical analysis are laborious, expensive, and only feasible for a limited land area. With the advent of remote sensing technologies like multi/hyperspectral imaging, it is now possible to predict soil properties non-invasive and cost-effectively for a large expanse of bare land. Recent research shows the possibility of predicting those soil contents from a wide range of hyperspectral data using good prediction algorithms. However, these kinds of hyperspectral sensors are expensive and not widely available. Therefore, this paper investigates different machine and deep learning techniques to predict soil nutrient properties using only the red (R), green (G), and blue (B) bands data to propose a suitable machine/deep learning model that can be used as a rapid soil test. Another objective of this research is to observe and compare the prediction accuracy in three cases i. hyperspectral band ii. full spectrum of the visual band, and iii. three-channel of RGB band and provide a guideline to the user on which spectrum information they should use to predict those soil properties. The outcome of this research helps to develop a mobile application that is easy to use for a quick soil test. This research also explores learning-based algorithms with significant feature combinations and their performance comparisons in predicting soil properties from visual band data. For this, we also explore the impact of dimensional reduction (i.e., principal component analysis) and transformations (i.e., empirical mode decomposition) of features. The results show that the proposed model can comparably predict the soil contents from the three-channel RGB data.
- Authors: Datta, Dristi , Paul, Manoranjan , Murshed, Manzur , Teng, Shyh Wei , Schmidtke, Leigh
- Date: 2023
- Type: Text , Journal article
- Relation: Environments Vol. 10, no. 5 (2023), p. 77
- Full Text:
- Reviewed:
- Description: Estimating various properties of soil, including moisture, carbon, and nitrogen, is crucial for studying their correlation with plant health and food production. However, conventional methods such as oven-drying and chemical analysis are laborious, expensive, and only feasible for a limited land area. With the advent of remote sensing technologies like multi/hyperspectral imaging, it is now possible to predict soil properties non-invasive and cost-effectively for a large expanse of bare land. Recent research shows the possibility of predicting those soil contents from a wide range of hyperspectral data using good prediction algorithms. However, these kinds of hyperspectral sensors are expensive and not widely available. Therefore, this paper investigates different machine and deep learning techniques to predict soil nutrient properties using only the red (R), green (G), and blue (B) bands data to propose a suitable machine/deep learning model that can be used as a rapid soil test. Another objective of this research is to observe and compare the prediction accuracy in three cases i. hyperspectral band ii. full spectrum of the visual band, and iii. three-channel of RGB band and provide a guideline to the user on which spectrum information they should use to predict those soil properties. The outcome of this research helps to develop a mobile application that is easy to use for a quick soil test. This research also explores learning-based algorithms with significant feature combinations and their performance comparisons in predicting soil properties from visual band data. For this, we also explore the impact of dimensional reduction (i.e., principal component analysis) and transformations (i.e., empirical mode decomposition) of features. The results show that the proposed model can comparably predict the soil contents from the three-channel RGB data.
Impact of traditional and embedded image denoising on CNN-based deep learning
- Kaur, Roopdeep, Karmakar, Gour, Imran, Muhammad
- Authors: Kaur, Roopdeep , Karmakar, Gour , Imran, Muhammad
- Date: 2023
- Type: Text , Journal article
- Relation: Applied sciences Vol. 13, no. 20 (2023), p.
- Full Text:
- Reviewed:
- Description: In digital image processing, filtering noise is an important step for reconstructing a high-quality image for further processing such as object segmentation, object detection, and object recognition. Various image-denoising approaches, including median, Gaussian, and bilateral filters, are available in the literature. Since convolutional neural networks (CNN) are able to directly learn complex patterns and features from data, they have become a popular choice for image-denoising tasks. As a result of their ability to learn and adapt to various denoising scenarios, CNNs are powerful tools for image denoising. Some deep learning techniques such as CNN incorporate denoising strategies directly into the CNN model layers. A primary limitation of these methods is their necessity to resize images to a consistent size. This resizing can result in a loss of vital image details, which might compromise CNN’s effectiveness. Because of this issue, we utilize a traditional denoising method as a preliminary step for noise reduction before applying CNN. To our knowledge, a comparative performance study of CNN using traditional and embedded denoising against a baseline approach (without denoising) is yet to be performed. To analyze the impact of denoising on the CNN performance, in this paper, firstly, we filter the noise from the images using traditional means of denoising method before their use in the CNN model. Secondly, we embed a denoising layer in the CNN model. To validate the performance of image denoising, we performed extensive experiments for both traffic sign and object recognition datasets. To decide whether denoising will be adopted and to decide on the type of filter to be used, we also present an approach exploiting the peak-signal-to-noise-ratio (PSNRs) distribution of images. Both CNN accuracy and PSNRs distribution are used to evaluate the effectiveness of the denoising approaches. As expected, the results vary with the type of filter, impact, and dataset used in both traditional and embedded denoising approaches. However, traditional denoising shows better accuracy, while embedded denoising shows lower computational time for most of the cases. Overall, this comparative study gives insights into whether denoising will be adopted in various CNN-based image analyses, including autonomous driving, animal detection, and facial recognition.
- Authors: Kaur, Roopdeep , Karmakar, Gour , Imran, Muhammad
- Date: 2023
- Type: Text , Journal article
- Relation: Applied sciences Vol. 13, no. 20 (2023), p.
- Full Text:
- Reviewed:
- Description: In digital image processing, filtering noise is an important step for reconstructing a high-quality image for further processing such as object segmentation, object detection, and object recognition. Various image-denoising approaches, including median, Gaussian, and bilateral filters, are available in the literature. Since convolutional neural networks (CNN) are able to directly learn complex patterns and features from data, they have become a popular choice for image-denoising tasks. As a result of their ability to learn and adapt to various denoising scenarios, CNNs are powerful tools for image denoising. Some deep learning techniques such as CNN incorporate denoising strategies directly into the CNN model layers. A primary limitation of these methods is their necessity to resize images to a consistent size. This resizing can result in a loss of vital image details, which might compromise CNN’s effectiveness. Because of this issue, we utilize a traditional denoising method as a preliminary step for noise reduction before applying CNN. To our knowledge, a comparative performance study of CNN using traditional and embedded denoising against a baseline approach (without denoising) is yet to be performed. To analyze the impact of denoising on the CNN performance, in this paper, firstly, we filter the noise from the images using traditional means of denoising method before their use in the CNN model. Secondly, we embed a denoising layer in the CNN model. To validate the performance of image denoising, we performed extensive experiments for both traffic sign and object recognition datasets. To decide whether denoising will be adopted and to decide on the type of filter to be used, we also present an approach exploiting the peak-signal-to-noise-ratio (PSNRs) distribution of images. Both CNN accuracy and PSNRs distribution are used to evaluate the effectiveness of the denoising approaches. As expected, the results vary with the type of filter, impact, and dataset used in both traditional and embedded denoising approaches. However, traditional denoising shows better accuracy, while embedded denoising shows lower computational time for most of the cases. Overall, this comparative study gives insights into whether denoising will be adopted in various CNN-based image analyses, including autonomous driving, animal detection, and facial recognition.
A micro-level compensation-based cost model for resource allocation in a fog environment
- Battula, Sudheer, Garg, Saurabh, Naha, Ranesh, Thulasiraman, Parimala, Thulasiram, Ruppa
- Authors: Battula, Sudheer , Garg, Saurabh , Naha, Ranesh , Thulasiraman, Parimala , Thulasiram, Ruppa
- Date: 2019
- Type: Text , Journal article
- Relation: Sensors Vol. 19, no. 13 (2019), p. 2954
- Full Text:
- Reviewed:
- Description: Fog computing aims to support applications requiring low latency and high scalability by using resources at the edge level. In general, fog computing comprises several autonomous mobile or static devices that share their idle resources to run different services. The providers of these devices also need to be compensated based on their device usage. In any fog-based resource-allocation problem, both cost and performance need to be considered for generating an efficient resource-allocation plan. Estimating the cost of using fog devices prior to the resource allocation helps to minimize the cost and maximize the performance of the system. In the fog computing domain, recent research works have proposed various resource-allocation algorithms without considering the compensation to resource providers and the cost estimation of the fog resources. Moreover, the existing cost models in similar paradigms such as in the cloud are not suitable for fog environments as the scaling of different autonomous resources with heterogeneity and variety of offerings is much more complicated. To fill this gap, this study first proposes a micro-level compensation cost model and then proposes a new resource-allocation method based on the cost model, which benefits both providers and users. Experimental results show that the proposed algorithm ensures better resource-allocation performance and lowers application processing costs when compared to the existing best-fit algorithm.
- Authors: Battula, Sudheer , Garg, Saurabh , Naha, Ranesh , Thulasiraman, Parimala , Thulasiram, Ruppa
- Date: 2019
- Type: Text , Journal article
- Relation: Sensors Vol. 19, no. 13 (2019), p. 2954
- Full Text:
- Reviewed:
- Description: Fog computing aims to support applications requiring low latency and high scalability by using resources at the edge level. In general, fog computing comprises several autonomous mobile or static devices that share their idle resources to run different services. The providers of these devices also need to be compensated based on their device usage. In any fog-based resource-allocation problem, both cost and performance need to be considered for generating an efficient resource-allocation plan. Estimating the cost of using fog devices prior to the resource allocation helps to minimize the cost and maximize the performance of the system. In the fog computing domain, recent research works have proposed various resource-allocation algorithms without considering the compensation to resource providers and the cost estimation of the fog resources. Moreover, the existing cost models in similar paradigms such as in the cloud are not suitable for fog environments as the scaling of different autonomous resources with heterogeneity and variety of offerings is much more complicated. To fill this gap, this study first proposes a micro-level compensation cost model and then proposes a new resource-allocation method based on the cost model, which benefits both providers and users. Experimental results show that the proposed algorithm ensures better resource-allocation performance and lowers application processing costs when compared to the existing best-fit algorithm.
- «
- ‹
- 1
- ›
- »