检索结果-内蒙古大学图书馆

optimization-based methodology for training set selection to synthesize composite correlation filters for face recognition

引用

SIGNAL PROCESSING-IMAGE COMMUNICATION 2016年 43卷 54-67页

作者： Santiago-Ramirez, Everardo Gonzalez-Fraga, Jose A. Gutierrez, Everardo Alvarez-Xochihua, Omar Univ Autonoma Baja California Fac Ciencias Carretera Transpeninsular Ensenada Tijuana3917 Ensenada 22860 Baja California Mexico Inst Politecn Nacl CITEDI Ave Parque 1310 Tijuana 22510 Baja California Mexico

Face recognition has been addressed with pattern recognition techniques such as composite correlation filters. These filters are synthesized from training sets which are representative of facial classes. For this reason, the filter performance depends greatly on the appropriate selection of the training set. This set can be selected either by a filter designer or by a conventional method. This paper presents an optimization-based methodology for the automatic selection of the training set. Given an optimization algorithm, the proposed methodology uses its main mechanics to iteratively examine a given set of available images in order to find the best subset for the training set. To this end, three objective functions are proposed as optimization criteria for training set selection. The proposed methodology was evaluated by undertaking face recognition under variable illumination and facial expressions. Four optimization algorithms and three composite correlation filters were used to test the proposed methodology. The Maximum Average Correlation Height filter designed by Grey Wolf Optimizer obtained the best performance under homogeneous illumination and facial expressions, while the Unconstrained Nonlinear Composite Filter designed by either Grey Wolf Optimizer or (1+1)-Evolution Strategy obtained the best performance under variable illumination. The proposed methodology selects training sets for the synthesis of composite filters with competitive results comparable to the results reported in the face recognition literature. (C) 2016 Elsevier B.V. All rights reserved.

关键词： Training set selection Face recognition Composite correlation filter optimization algorithm Pattern recognition

来源：评论

学校读者我要写书评

暂无评论

Multiobjective optimization Models for Locating Vehicle Inspection Stations Subject to Stochastic Demand, Varying Velocity and Regional Constraints

引用

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS 2016年第7期17卷 1978-1987页

作者： Tian, Guangdong Zhou, MengChu Li, Peigen Zhang, Chaoyong Jia, Hongfei Jilin Univ Coll Transportat Changchun 130022 Peoples R China Huazhong Univ Sci & Technol State Key Lab Digital Mfg Equipment & Technol Wuhan 430074 Peoples R China Northeast Forestry Univ Coll Transportat Harbin 150040 Peoples R China Macau Univ Sci & Technol Inst Syst Engn Macau 999078 Peoples R China New Jersey Inst Technol Dept Elect & Comp Engn Newark NJ 07102 USA

Deciding an optimal location of a transportation facility and automotive service enterprise is an interesting and important issue in the area of facility location allocation (FLA). In practice, some factors, i.e., customer demands, allocations, and locations of customers and facilities, are changing, and thus, it features with uncertainty. To account for this uncertainty, some researchers have addressed the stochastic time and cost issues of FLA. A new FLA research issue arises when decision makers want to minimize the transportation time of customers and their transportation cost while ensuring customers to arrive at their desired destination within some specific time and cost. By taking the vehicle inspection station as a typical automotive service enterprise example, this paper presents a novel stochastic multiobjective optimization to address it. This work builds two practical stochastic multiobjective programs subject to stochastic demand, varying velocity, and regional constraints. A hybrid intelligent algorithm integrating stochastic simulation and multiobjective teaching-learning-based optimization algorithm is proposed to solve the proposed programs. This approach is applied to a real-world location problem of a vehicle inspection station in Fushun, China. The results show that this is able to produce satisfactory Pareto solutions for an actual vehicle inspection station location problem.

关键词： Transportation facility location optimization algorithm modeling and simulation

来源：评论

学校读者我要写书评

暂无评论

Modelling and optimization of the velocity profiles at the draft tube inlet of a Francis turbine within an operating range

引用

JOURNAL OF HYDRAULIC RESEARCH 2016年第1期54卷 74-89页

作者： Ciocan, Tiberiu Susan-Resiga, Romeo F. Muntean, Sebastian Politehn Univ Timisoara Res Ctr Engn Syst Complex Fluids Timisoara Romania Politehn Univ Timisoara Hydraul Machinery Dept Timisoara Romania Romanian Acad Ctr Adv Res Engn Sci Timisoara Branch Timisoara Romania

Operating hydraulic turbines over an extended range of discharge values requires an optimized runner that minimizes the weighted-average draft tube losses. It is shown that the draft tube losses can be minimized before actually designing the (new or refurbished) runner blades by optimizing the runner outlet flow. For this, a new runner surrogate model is introduced, where the relative flow angle is expressed via the so-called swirl-free velocity profile (represented with two parameters). Given a swirl-free velocity profile, as well as a turbine discharge value, the model provides swirling flow profiles consistent with the hydraulic turbine operation. With this swirling flow as inlet condition, the draft tube flow computation provides the relationship between the swirl-free velocity parameters and the hydraulic loss. It is shown that this approach can be used for optimizing the runner for a range of turbine operating regimes.

关键词： swirling flow draft tube CFD optimization algorithm hydraulic turbine

来源：评论

学校读者我要写书评

暂无评论

ON THE USE OF SEMANTIC AWARENESS TO LIMIT OVERFITTING IN GENETIC PROGRAMMING

ON THE USE OF SEMANTIC AWARENESS TO LIMIT OVERFITTING IN GEN...

引用

作者： Paul Joscha Englert Universidade Nova de Lisboa

学位级别：硕士

Nowadays, the use of creativity in business has been increasing drastically because it is important for the market to come up with new ways to find answers to the problems proposed by the users. There are several different creativity techniques that can be used in different contexts. One of the most important techniques is the SCAMPER technique, which is based in reorganizing, modifying, adding and eliminating information. An automated system will provide answers and solutions to creativity problems and contribute to minimize the cost of innovation in the companies. For that reason businesses will be able and willing to continue to use creativity when innovating. The aim of this thesis is therefore to design an architecture system for a creative information system based on an automated system that relies on the SCAMPER creativity technique.

关键词： Machine Learning Statistics Computational Intelligence Genetic Programming Genetic algorithm Evolutionary algorithm optimization algorithm optimization Problem Overfitting Semantic Awareness Semantic Repulser Multi-Objective System Semantic Signature

来源：评论

学校读者我要写书评

暂无评论

Real-time combined heat and power operational strategy using a hierarchical optimization algorithm

引用

PROCEEDINGS OF THE INSTITUTION OF MECHANICAL ENGINEERS PART A-JOURNAL OF POWER AND ENERGY 2011年第A4期225卷 403-412页

作者： Yun, K. Cho, H. Luck, R. Mago, P. J. Mississippi State Univ Dept Mech Engn Mississippi State MS 39762 USA Pacific NW Natl Lab Richland WA 99352 USA

Existing attempts to optimize the operation of combined heat and power (CHP) systems for building applications have two major limitations: the electrical and thermal loads are obtained from historical weather profiles;and the CHP system models ignore transient responses by using constant equipment efficiencies. This article considers the transient response of a building combined with a hierarchical CHP optimal control algorithm to obtain a real-time integrated system that uses the most recent weather and electric load information. This is accomplished by running concurrent simulations of two transient building models. The first transient building model uses current as well as forecast input information to obtain short-term predictions of the thermal and electric building loads. The predictions are then used by an optimization algorithm (i.e. a hierarchical controller that decides the amount of fuel and of electrical energy to be allocated at the current time step). In a simulation, the actual physical building is not available and, hence, to simulate a real-time environment, a second, building model with similar but not identical input loads are used to represent the actual building. A state-variable feedback loop is completed at the beginning of each time step by copying (i.e. measuring, the state variable from the actual building and restarting the predictive model using these 'measured' values as initial conditions). The simulation environment presented in this article features non-linear effects such as the dependence of the heat exchanger effectiveness on their operating conditions. The results indicate that the CHP engine operation dictated by the proposed hierarchical controller with uncertain weather conditions has the potential to yield significant savings when compared with conventional systems using current values of electricity and fuel prices.

关键词： combined heat and power real-time optimization optimization algorithm

来源：评论

学校读者我要写书评

暂无评论

Cognitive radar ambiguity function optimization for unimodular sequence

引用

EURASIP JOURNAL ON ADVANCES IN SIGNAL PROCESSING 2016年第1期2016卷 1页

作者： Zhang, Jindong Qiu, Xiaoyan Shi, Changli Wu, Yue Nanjing Univ Aeronaut & Astronaut Coll Elect & Informat Engn Nanjing 210016 Jiangsu Peoples R China

An important characteristic of a cognitive radar is the capability to adjust its transmitted waveform to adapt to the radar environment. The adaptation of the transmit waveform requires an effective framework to synthesize waveforms sharing a desired ambiguity function (AF). With the volume-invariant property of AF, the integrated sidelobe level (ISL) can only be minimized in a certain area on the time delay and Doppler frequency shift plane. In this paper, we propose a new algorithm for unimodular sequence to minimize the ISL of an AF in a certain area based on the phase-only conjugate gradient and phase-only Newton's method. For improving detection performance of a moving target detecting (MTD) radar system, slow-time ambiguity function (STAF) is defined, and the proposed algorithm is presented to optimize the range-Doppler response. We also devise a cognitive approach for a MTD radar by adaptively altering its sidelobe distribution of STAF. At the simulation stage, the performance of the proposed algorithm is assessed to show their capability to properly shape the AF and STAF of the transmitted waveform.

关键词： Cognitive radar Unimodular sequence synthesis optimization algorithm Ambiguity function Slow-time ambiguity function

来源：评论

学校读者我要写书评

暂无评论

Multidisciplinary optimization Under High-Dimensional Uncertainty for Small Satellite System Design

引用

AIAA JOURNAL 2016年第5期54卷 1732-1741页

作者： Hu, Xingzhi Chen, Xiaoqian Lattarulo, Valerio Parks, Geoffrey T. Natl Univ Def Technol Coll Aerosp Sci & Engn Changsha 410073 Hunan Peoples R China Univ Cambridge Dept Engn Cambridge CB2 1PZ England

Reliability-based robust design optimization of modern small satellites has recently been receiving much attention. The role of multidisciplinary design optimization considering system uncertainty is increasingly being recognized in improving satellite performance, safety, and reliability. However, high-dimensional uncertainty hampers the efficiency and accuracy of reliability-based robust design optimization. In this study, a novel multidisciplinary optimization methodology suitable for high-dimensional uncertainty is established and verified. The in-loop uncertainty quantification employing active subspace identification makes the uncertainty propagation and management much easier. Based on the discovered active subspace and a system decoupling strategy, a multiobjective alliance algorithm is presented for the complex constrained design optimization. Typical uncertain factors in small satellites are characterized and interdisciplinary relations are analyzed. A standard test example of speed reducer design first proves the methodology, and then an Earth observation satellite fully illustrates the efficacy. Through 23-dimensional uncertainty represented by one dimension, the multiobjective alliance algorithm gives well-distributed Pareto-optimal (PO) solutions, providing a fruitful reference for decision makers compromising different design targets, like satellite mass, total cost, and ground resolution. The proposed approach is proved to be fairly accurate and adaptable, which can be widely applied to uncertainty-based multidisciplinary design optimization (UMDO) of satellite systems.

关键词： Low Earth Orbit Satellites MDO EOS Robust Design Finite Difference Approximation Ultimate Tensile Strength Surrogate Model Small Satellite Cost Model optimization algorithm European Space Agency

来源：评论

学校读者我要写书评

暂无评论

Control of gold nano-domes to design an all-optical switch by simultaneously binary and continuous operated optimization

引用

JOURNAL OF COMPUTATIONAL ELECTRONICS 2016年第2期15卷 569-576页

作者： Keshavarzi, Rasoul Emami, Farzin Islamic Azad Univ Omidieh Branch Young Researchers & Elite Club Omidieh Iran

This paper presents an efficient evolutionary method to optimize an array of plasmonic nano-domes in order to design an optical switch. In the proposed switch, the optical switch has been excited by two monochromatic incident plan-waves with the same frequency and two polar angles "" and "". When only the signal with is applied, the incident wave is transmitted and when both signals are applied to the switch simultaneously, coherent perfect absorption (CPA) occurs and the two incident waves are suppressed. Therefore, the signal with acts as control signal. Since the CPA efficiency depends strongly on the number of plasmonic nano-domes, on their location and geometric characterization, a new efficient optimization method named "fuzzy adaptive modified particle swarm optimization" (FAMPSO) is proposed to design an optimized array of the plasmonic nano-domes in order to achieve the maximum absorption coefficient in the off-state and the minimum absorption coefficient in the on-state. In the FAMPSO algorithm a group of birds controlling the presence ('1') or the absence ('0') and geometric characterization of nano domes in the array.

关键词： Plasmonic switch Plasmonic nano particles optimization algorithm

来源：评论

学校读者我要写书评

暂无评论

Synthesis of hexagonal planar array using swarm-based optimization algorithms

引用

INTERNATIONAL JOURNAL OF MICROWAVE AND WIRELESS TECHNOLOGIES 2015年第2期7卷 151-160页

作者： Chatterjee, Anirban Mandal, Debasis Natl Inst Technol Goa Dept Elect & Commun Engn Ponda Goa India Bengal Coll Engn & Technol Dept Elect & Commun Engn Durgapur India

Nature inspired optimization algorithms, namely artificial bee colony (ABC) optimization and firefly algorithm (FA), have been applied to synthesize beam patterns of a hexagonal planar array of isotropic elements. Two different cases, comprising two different beam patterns of a pencil beam and a square footprint pattern over a bounded region with lower peak sidelobe levels are presented. The pencil beam is generated by thinning the uniformly excited array and the square footprint pattern is generated by imposing optimum amplitudes, phases, and their corresponding states ("on"/"off") to the array elements. The optimum values of the parameters for both the cases are computed using ABC and FA individually, and the superiority of FA over ABC for the proposed problem in terms of computing solutions for both the cases is established.

关键词： optimization algorithm Artificial bee colony (ABC) algorithm Firefly algorithm (FA) Hexagonal planar array Sidelobe level (SLL) Square footprint Thinning

来源：评论

学校读者我要写书评

暂无评论

Performance Enhancement of Global optimization-Based Gas Turbine Fault Diagnosis Systems

引用

JOURNAL OF PROPULSION AND POWER 2016年第1期32卷 214-224页

作者： Mohammadi, Ehsan Montazeri-Gh, Morteza Iran Univ Sci & Technol Sch Mech Engn Syst Simulat & Control Lab Tehran *** Iran

Fault detection and identification of gas turbines is a crucial process for providing engine safe operation and decreasing the maintenance costs. In studies conducted in the field of global optimization-based gas turbine fault diagnosis, the genetic algorithm as the most well-known evolutionary optimization algorithm is usually employed to identify the engine health parameters. However, because of the evolutionary and stochastic nature of this algorithm, the genetic-algorithm-based diagnosis usually suffers from computational burden and reliability. To mitigate this problem, in the present work, a comparative study has been performed on the global optimization-based gas turbine fault diagnosis, and it is shown that an innovative hybrid optimization algorithm as a fault detection and identification system can significantly enhance the performance of the conventional optimization-based diagnosis systems, even in the presence of measurement noise. The results obtained indicate that the fault detection and identification system based on the hybrid invasive weed optimization/particle swarm optimization algorithm outperforms all the examined diagnosis systems (i.e., the genetic-algorithm-based, particle-swarm-optimization-based, and invasive weed-optimization-based fault detection and identification system) in terms of accuracy, reliability, and especially computational cost. The results demonstrate that the genetic-algorithm-based fault detection and identification system showed the weakest performance among all the examined diagnosis systems.

关键词： Gas Turbine Particle Swarm optimization optimization algorithm Fault Detection Genetic algorithm Axial Compressors Artificial Neural Network Bayesian Belief Networks Influence Coefficients Gas Path Analysis

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：