Videos have vital applications in numerous real-time aspects such as teaching, learning, communic... more Videos have vital applications in numerous real-time aspects such as teaching, learning, communication, computer vision, and medicine. Typically, video registration is required to describe a part of the scene/object in the video frame or to localize an object in the frame relative to a fixed reference system. The semilocal transformation generated by the B-splines registration was solved using demons algorithm. Thus, the current study is concerned with demons algorithm–based image registration for a fully local transformation model. The demons registration is optimized using the firefly algorithm (FA) to optimize the velocity-smoothing kernels of the demons registration considering the correlation coefficient as a fitness function. Afterward, the proposed system performance using demons algorithm–based FA is compared to the particle swarm optimization (PSO). The experimental results proved that the proposed system-based FA achieved a correlation value of 0.6108 compared to demons registration with default parameters that provided 0.4468. Additionally, the FA-based optimization framework was more stable and produced superior results than the PSO-based optimization framework. In addition, the FA algorithm converged faster than the PSO one.
This work presents an application of bio-inspired flower pollination algorithm (FPA) for tuning p... more This work presents an application of bio-inspired flower pollination algorithm (FPA) for tuning proportional–integral–derivative (PID) controller in load frequency control (LFC) of multi-area interconnected power system. The investigated power system comprises of three equal thermal power systems with appropriate PID controller. The controller gain [proportional gain (Kp), integral gain (Ki) and derivative gain (Kd)] values are tuned by using the FPA algorithm with one percent step load perturbation in area 1 (1 % SLP). The integral square error (ISE) is considered the objective function for the FPA. The supremacy performance of proposed algorithm for optimized PID controller is proved by comparing the results with genetic algorithm (GA) and particle swarm optimization (PSO)-based PID controller under the same investigated power system. In addition, the controller robustness is studied by considering appropriate generate rate constraint with nonlinearity in all areas. The result cumulative performance comparisons established that FPA-PID controller exhibit better performance compared to performances of GA-PID and PSO-PID controller-based power system with and without nonlinearity effect.
Authentication of medical data and reports is a significant issue for preparing, editing, and tra... more Authentication of medical data and reports is a significant issue for preparing, editing, and transmitting the patient report. Medical information transfer requires communication technologies in order to offer health care services. This achieves mutual availability of therapeutic case studies among different diagnostic hospitals or centers which are very widespread. One key for accomplishing information security for reliable transmission purposes is digital watermarking through the embedding of hidden information, a hospital logo, or secret information within the medical images. The present work proposes a novel authentication system for robust biomedical content, which embeds multiple hospitals’ logos or multiple electronic patient records within the medical image. A hybrid approach for watermarking based on discrete cosine transformation, discrete wavelet transformation, and singular value decomposition teamed with optimization algorithms is proposed. The optimization algorithms used include a genetic algorithm, a quantum-inspired genetic algorithm, and a quantum-inspired evolutionary algorithm. These algorithms are applied to discover the optimal possible scaling factors for embedding. Three different medical sets are used to examine the proposed approach, including microscopic, ultrasound, and magnetic resonance imaging images. The results show that the proposed method established an accurate authentication system for information exchange with high robustness, imperceptibility, and payload. The results prove the dominance of both the quantum-inspired evolutionary algorithm and quantum-inspired genetic algorithm over the genetic algorithm.
The primary goal of this article is to design and implement a secondary controller with which to ... more The primary goal of this article is to design and implement a secondary controller with which to control the system frequency in a networked microgrid system. The proposed power system comprises of Renewable energy sources (RESs), energy-storing units (ESUs), and synchronous generator. RESs include photovoltaic (PV) and wind turbine generator (WTG) units. The ESU is composed of a flywheel and a battery. Because renewable energy sources are not constant in nature, their values fluctuate from time to time, causing an effect on system frequency and power flow variation in the tie line. The nonlinear output from the RESs is balanced with the support of ESUs. In order to address this situation, a proportional integral derivative (PID) controller based on the Mayfly algorithm (MA) is proposed and built. Comparing the responses of controllers based on the genetic algorithm (GA), differential evolution (DE), and particle swarm optimization (PSO) technique-optimized to demonstrate the superi...
2016 International Conference on Computer, Electrical & Communication Engineering (ICCECE)
In this paper, we propose the nonlinear multimodal optimization problem of low pass FIR filter de... more In this paper, we propose the nonlinear multimodal optimization problem of low pass FIR filter design using three conventional optimization algorithms. The efficacy of the proposed method was compared with the traditional approach of filter design Parks McClellan algorithm as reference. In optimization algorithms mean square error based cost function was used as the fitness function. It is seen was observed that the BAT algorithm statistically outperforms Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) in terms of stopband attenuation characteristics and ripple performance of the designed filter.
Load Frequency Control (LFC) is an important issue in power system to maintain power system stabi... more Load Frequency Control (LFC) is an important issue in power system to maintain power system stability and quality of generated power supply during sudden load demand period. In order to overcome this issue, power systems are interconnected and secondary controllers are introduced to regulate the power system parameters within a specified limit during sudden load demand period. In this present work, three area single stage reheat thermal power systems are interconnected and each area comprises governor unit, reheated unit, turbine unit, Governor Dead Band (GDB), Generation Rate Constraint (GRC) non-linear components and boiler dynamics effect. One Percent (1%) Step Load Perturbation (SLP) is considered in thermal area 1 of the investigated power system. The Proportional-Integral-Derivative (PID) controller is introduced as a secondary controller. Since tuning of the controller gain values play a vital role, evolutionary algorithms are introduced to tune the controller gain values. In the current work, the Genetic Algorithm (GA), Quantum Inspired Genetic Algorithm (QIGA) and Quantum Inspired Evolutionary Algorithm (QIEA) are proposed for tuning of controller gain values. The cumulative comparisons of the simulation result are clearly reported that QIGA and QIEA are more superior to the GA based PID controller performance in the same investigated power system in terms of time domain specification parameters.
AbstractIn this paper we proposed a method of image encoding technique in a new formulated way. ... more AbstractIn this paper we proposed a method of image encoding technique in a new formulated way. During transmission image is encoded in two phases for the enhancement of security. In the first phase encoding is done using Walsh table and in the second phase ...
The Unmanned Aerial Vehicles (UAV) are widely used for capturing images in border area surveillan... more The Unmanned Aerial Vehicles (UAV) are widely used for capturing images in border area surveillance, disaster intensity monitoring, etc. An aerial photograph offers a permanent recording solution as well. But rapid weather change, low quality image capturing equipments results in low/poor contrast images during image acquisition by Autonomous UAV. In this current study, a well-known meta-heuristic technique, namely, Firefly Algorithm (FA) is reported to enhance aerial images taken by a Mini Unmanned Aerial Vehicle (MUAV) via optimizing the value of certain parameters. These parameters have a wide range as used in the Log Transformation for image enhancement. The entropy and edge information of the images is used as an objective criterion for evaluating the image enhancement of the proposed system. Inconsistent with the objective criterion, the FA is used to optimize the parameters employed in the objective function that accomplishes the superlative enhanced image. A low-light imagin...
Videos have vital applications in numerous real-time aspects such as teaching, learning, communic... more Videos have vital applications in numerous real-time aspects such as teaching, learning, communication, computer vision, and medicine. Typically, video registration is required to describe a part of the scene/object in the video frame or to localize an object in the frame relative to a fixed reference system. The semilocal transformation generated by the B-splines registration was solved using demons algorithm. Thus, the current study is concerned with demons algorithm–based image registration for a fully local transformation model. The demons registration is optimized using the firefly algorithm (FA) to optimize the velocity-smoothing kernels of the demons registration considering the correlation coefficient as a fitness function. Afterward, the proposed system performance using demons algorithm–based FA is compared to the particle swarm optimization (PSO). The experimental results proved that the proposed system-based FA achieved a correlation value of 0.6108 compared to demons registration with default parameters that provided 0.4468. Additionally, the FA-based optimization framework was more stable and produced superior results than the PSO-based optimization framework. In addition, the FA algorithm converged faster than the PSO one.
This work presents an application of bio-inspired flower pollination algorithm (FPA) for tuning p... more This work presents an application of bio-inspired flower pollination algorithm (FPA) for tuning proportional–integral–derivative (PID) controller in load frequency control (LFC) of multi-area interconnected power system. The investigated power system comprises of three equal thermal power systems with appropriate PID controller. The controller gain [proportional gain (Kp), integral gain (Ki) and derivative gain (Kd)] values are tuned by using the FPA algorithm with one percent step load perturbation in area 1 (1 % SLP). The integral square error (ISE) is considered the objective function for the FPA. The supremacy performance of proposed algorithm for optimized PID controller is proved by comparing the results with genetic algorithm (GA) and particle swarm optimization (PSO)-based PID controller under the same investigated power system. In addition, the controller robustness is studied by considering appropriate generate rate constraint with nonlinearity in all areas. The result cumulative performance comparisons established that FPA-PID controller exhibit better performance compared to performances of GA-PID and PSO-PID controller-based power system with and without nonlinearity effect.
Authentication of medical data and reports is a significant issue for preparing, editing, and tra... more Authentication of medical data and reports is a significant issue for preparing, editing, and transmitting the patient report. Medical information transfer requires communication technologies in order to offer health care services. This achieves mutual availability of therapeutic case studies among different diagnostic hospitals or centers which are very widespread. One key for accomplishing information security for reliable transmission purposes is digital watermarking through the embedding of hidden information, a hospital logo, or secret information within the medical images. The present work proposes a novel authentication system for robust biomedical content, which embeds multiple hospitals’ logos or multiple electronic patient records within the medical image. A hybrid approach for watermarking based on discrete cosine transformation, discrete wavelet transformation, and singular value decomposition teamed with optimization algorithms is proposed. The optimization algorithms used include a genetic algorithm, a quantum-inspired genetic algorithm, and a quantum-inspired evolutionary algorithm. These algorithms are applied to discover the optimal possible scaling factors for embedding. Three different medical sets are used to examine the proposed approach, including microscopic, ultrasound, and magnetic resonance imaging images. The results show that the proposed method established an accurate authentication system for information exchange with high robustness, imperceptibility, and payload. The results prove the dominance of both the quantum-inspired evolutionary algorithm and quantum-inspired genetic algorithm over the genetic algorithm.
The primary goal of this article is to design and implement a secondary controller with which to ... more The primary goal of this article is to design and implement a secondary controller with which to control the system frequency in a networked microgrid system. The proposed power system comprises of Renewable energy sources (RESs), energy-storing units (ESUs), and synchronous generator. RESs include photovoltaic (PV) and wind turbine generator (WTG) units. The ESU is composed of a flywheel and a battery. Because renewable energy sources are not constant in nature, their values fluctuate from time to time, causing an effect on system frequency and power flow variation in the tie line. The nonlinear output from the RESs is balanced with the support of ESUs. In order to address this situation, a proportional integral derivative (PID) controller based on the Mayfly algorithm (MA) is proposed and built. Comparing the responses of controllers based on the genetic algorithm (GA), differential evolution (DE), and particle swarm optimization (PSO) technique-optimized to demonstrate the superi...
2016 International Conference on Computer, Electrical & Communication Engineering (ICCECE)
In this paper, we propose the nonlinear multimodal optimization problem of low pass FIR filter de... more In this paper, we propose the nonlinear multimodal optimization problem of low pass FIR filter design using three conventional optimization algorithms. The efficacy of the proposed method was compared with the traditional approach of filter design Parks McClellan algorithm as reference. In optimization algorithms mean square error based cost function was used as the fitness function. It is seen was observed that the BAT algorithm statistically outperforms Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) in terms of stopband attenuation characteristics and ripple performance of the designed filter.
Load Frequency Control (LFC) is an important issue in power system to maintain power system stabi... more Load Frequency Control (LFC) is an important issue in power system to maintain power system stability and quality of generated power supply during sudden load demand period. In order to overcome this issue, power systems are interconnected and secondary controllers are introduced to regulate the power system parameters within a specified limit during sudden load demand period. In this present work, three area single stage reheat thermal power systems are interconnected and each area comprises governor unit, reheated unit, turbine unit, Governor Dead Band (GDB), Generation Rate Constraint (GRC) non-linear components and boiler dynamics effect. One Percent (1%) Step Load Perturbation (SLP) is considered in thermal area 1 of the investigated power system. The Proportional-Integral-Derivative (PID) controller is introduced as a secondary controller. Since tuning of the controller gain values play a vital role, evolutionary algorithms are introduced to tune the controller gain values. In the current work, the Genetic Algorithm (GA), Quantum Inspired Genetic Algorithm (QIGA) and Quantum Inspired Evolutionary Algorithm (QIEA) are proposed for tuning of controller gain values. The cumulative comparisons of the simulation result are clearly reported that QIGA and QIEA are more superior to the GA based PID controller performance in the same investigated power system in terms of time domain specification parameters.
AbstractIn this paper we proposed a method of image encoding technique in a new formulated way. ... more AbstractIn this paper we proposed a method of image encoding technique in a new formulated way. During transmission image is encoded in two phases for the enhancement of security. In the first phase encoding is done using Walsh table and in the second phase ...
The Unmanned Aerial Vehicles (UAV) are widely used for capturing images in border area surveillan... more The Unmanned Aerial Vehicles (UAV) are widely used for capturing images in border area surveillance, disaster intensity monitoring, etc. An aerial photograph offers a permanent recording solution as well. But rapid weather change, low quality image capturing equipments results in low/poor contrast images during image acquisition by Autonomous UAV. In this current study, a well-known meta-heuristic technique, namely, Firefly Algorithm (FA) is reported to enhance aerial images taken by a Mini Unmanned Aerial Vehicle (MUAV) via optimizing the value of certain parameters. These parameters have a wide range as used in the Log Transformation for image enhancement. The entropy and edge information of the images is used as an objective criterion for evaluating the image enhancement of the proposed system. Inconsistent with the objective criterion, the FA is used to optimize the parameters employed in the objective function that accomplishes the superlative enhanced image. A low-light imagin...
Uploads
Papers by Sourav Samanta