基于目标识别的光电多传感器信息融合技术

依靠多个光电传感器采集目标数据,需要解决各类数据信息的融合分析问题。基于这种认识,从特征提取、融合和识别等方面,对基于目标识别的光电多传感器信息融合技术展开分析,介绍了技术应用前景,指出需要采用不同算法完成不同光电传感器特征信号提取,同时,通过融合计算得到融合特征量,为目标识别提供信息数据支撑,确保准确识别目标。     

模糊集理论在多传感器信息融合中的应用

在多传感器信息融合系统中,融合系统处理的信息本质具有模糊性,而模糊集理论具有处理模糊问题和模糊推理的优势,因此,模糊集理论已被广泛应用在多传感器信息融合领域。描述的信息融合方法中,通过引入隶属函数的概念,对传感器的测量值进行模糊化处理;利用模糊综合评判原理把传感器的信息融合问题转化为模糊综合评判过程。通过仿真实验验证,这种信息融合方法计算量小、信息融合精度高。     

基于多传感器信息融合的控制系统研究

本文研究了多传感器组合系统的智能化数据融合方法。从数据处理算法、故障检测和系统可靠性等方面探讨了基于多传感器的控制系统,并提出了一种实用的数据处理算法和故障检测算法。     

目标识别中多传感器信息融合算法比较

近年来多传感器信息融合技术在目标识别领域得到了大量研究和快速发展. 介绍了多传感器信息融合目标识别的基本原理及其系统结构, 重点阐述了目标识别中的多传感器信息融合算法, 并对识别效果进行比较, 最后指出了该领域今后的发展趋势.     

多传感器信息融合目标识别系统设计

多传感器信息融合主要用于目标检测、定位、跟踪和识别。多传感器信息融合对来自不同信息源的信息进行分析与综合,产生被测对象的统一最佳估计,使信息的准确性、可靠性及完备性有明显提高。各个传感器所提供的信息一般是不完整的,即包含大量的不确定性。而证据理论能很好地表示不确定性,且推理形式简单,因而在信息融合方面起着重要的作用。本文采用模糊集合的隶属度函数构造证据理论中的基本概率赋值函数,使得证据理论应用于实际更加方便有效。该方法首先根据被跟踪目标数据库的信息构建每个属性的模糊逻辑图,然后以模糊集合的隶属度函数为基础计算每个属性的mass函数。最后用证据理论的合成规则对mass函数进行合成达到目标识别的目的。     

多传感器信息融合模糊控制模型设计

多传感器数据的来源众多,数据时间序列的特征随机性强,难以统一,导致其信息应用范围缩小。提出一种多传感器信息融合的模糊控制模型。运用不同映射模式描述多传感器信息融合状态空间,创建随机时段下测量空间矩阵,获得传感器信息采集时间序列特征。根据信息采集时间序列特征构建二级架构信息融合模型,第一级架构使用模糊控制算法划分输入-输出空间模糊区间,得到模糊规则并计算模糊规则相对信任度,利用模糊规则映射关联聚类信息,剔除传感器冗余数据。在此基础上使用智能粒子滤波法将多传感器信息传输至相应粒子滤波模块,代入遗传算法交叉与变异操作调整粒子权重,通过重采样保存高权值粒子,得到完整的多传感器信息融合结果。仿真结果表明,多传感器信息融合的最大能耗值为110 mJ,信息采集网络延迟为0.75 s,融合时间平均值为4.5 s,信息融合的误差值小于50 m,系统鲁棒性较强。     

基于信息熵的多传感器信息融合

多传感器信息融合是多源信息综合处理的一项新技术。从信息论的观点出发,导出多传感器信息融合系统中的冗余性与互补性的定量描述,分析了传感器冗余性与互补性特点,并从该角度出发,利用最小条件熵准则来解决多传感器信息融合中的目标识别问题,该方法的主要优点是可以充分有效利用多传感器信息,使融合系统满足获得的信息量最大。     

基于多传感器的智能机器人信息融合,控制结构和应用

智能机器人能最佳地、综合地使用多传感器信息,具有完成某一特写任务所需的完备信息,从而实现在各种复杂、动态的、不确定环境中的自主性、多传感器信息融合是指将经过集成处理的多传感器信息进行合成。形成对外部环境某一特征的一种表达方式,经过集成与融合的多传感器信息能完善地、精确地反映环境特征,而单一传感器只能获得环境特征的部分信息段,经过集成与融合的多传感器信息具有以下特征,信息冗余性、信息的互补性、信息的     

基于信息融合的多传感器侦察技术

本文介绍了多传感器侦察系统中信息融合技术的特点、层次、结构模型以及实现方法,提出了未来多传感器侦察系统信息融合技术应着力解决的问题。     

基于遗传算法的数据挖掘技术的研究

文章首先对数据挖掘进行了概述,阐明了什么是数据挖掘,为什么要数据挖掘,如何进行数据挖掘以及数据挖掘的主要过程,接着介绍了数据挖掘中的一个重要算法－遗传算法,文章对遗传算法的产生与发展以及主要理论等进行了简要的介绍,提出了基于遗传算法的关联规则的提取方法,文章还结合作者单位的智能型学生管理系统,给出了用遗传算法进行关联规则挖掘的实例,并讨论了遗传算法所面临的问题与挑战。     

一种基于遗传算法的兴趣规则挖掘算法

数据挖掘是在数据中发现隐藏的结构和模式。但发现的许多模式对用卢来说可能是已知的，从而使这些模式毫无意义，毫无兴趣性。文献中多强调分类规则的准确性和可理解性，但发现兴趣规则在数据挖掘算法中依然是一个令人生畏的挑战。本文采用一种遗传数据挖掘方法，在分类规则产生的同时对其兴趣性进行度量，直接产生兴趣规则。实验表明该方法是可行的、高效的。     

免疫遗传算法在图像分割中的应用

提出一种基于免疫系统的免疫记忆特性所改进的遗传算法。该算法方面在传统的遗传算法的初始种群中,加入了根据先验知识制成的疫苗,从而大大提高了算法的收敛速度;另一方面,对遗传算子中的选择算子也进行了改进,吸取了免疫系统中的克隆选择的优点,并且根据细胞的亲和力进行变异,进而提高了图像分割的速度。     

A grid-oriented genetic algorithm framework for bioinformatics

In this paper, we propose a framework for enabling for researchers of genetic algorithms (GAs) to easily develop GAs running on the Grid, named “Grid-Oriented Genetic algorithms (GOGAs)”, and actually “Gridify” a GA for estimating genetic networks, which is being developed by our group, in order to examine the usability of the proposed GOGA framework. We also evaluate the scalability of the “Gridified” GA by applying it to a five-gene genetic network estimation problem on a grid testbed constructed in our laboratory. Hiroaki Imade: He received his B.S. degree in the department of engineering from The University of Tokushima, Tokushima, Japan, in 2001. He received the M.S. degree in information systems from the Graduate School of Engineering, The University of Tokushima in 2003. He is now in Doctoral Course of Graduate School of Engineering, The University of Tokushima. His research interests include evolutionary computation. He currently researches a framework to easily develop the GOGA models which efficiently work on the grid. Ryohei Morishita: He received his B.S. degree in the department of engineering from The University of Tokushima, Tokushima, Japan, in 2002. He is now in Master Course of Graduate School of Engineering, The University of Tokushima, Tokushima. His research interest is evolutionary computation. He currently researches GA for estimating genetic networks. Isao Ono, Ph.D.: He received his B.S. degree from the Department of Control Engineering, Tokyo Institute of Technology, Tokyo, Japan, in 1994. He received Ph.D. of Engineering at Tokyo Institute of Technology, Yokohama, in 1997. He worked as a Research Fellow from 1997 to 1998 at Tokyo Institute of Technology, and at University of Tokushima, Tokushima, Japan, in 1998. He worked as a Lecturer from 1998 to 2001 at University of Tokushima. He is now Associate Professor at University of Tokushima. His research interests include evolutionary computation, scheduling, function optimization, optical design and bioinformatics. He is a member of JSAI, SCI, IPSJ and OSJ. Norihiko Ono, Ph.D.: He received his B.S. M.S. and Ph.D. of Engineering in 1979, 1981 and 1986, respectively, from Tokyo Institute of Technology. From 1986 to 1989, he was Research Associate at Faculty of Engineering, Hiroshima University. From 1989 to 1997, he was an associate professor at Faculty of Engineering, University of Tokushima. He was promoted to Professor in the Department of Information Science and Intelligent Systems in 1997. His current research interests include learning in multi-agent systems, autonomous agents, reinforcement learning and evolutionary algorithms. Masahiro Okamoto, Ph.D.: He is currently Professor of Graduate School of Systems Life Sciences, Kyushu University, Japan. He received his Ph.D. degree in Biochemistry from Kyushu University in 1981. His major research field is nonlinear numerical optimization and systems biology. His current research interests cover system identification of nonlinear complex systems by using evolutional computer algorithm of optimization, development of integrated simulator for analyzing nonlinear dynamics and design of fault-tolerant routing network by mimicking metabolic control system. He has more than 90 peer reviewed publications.     

一种基于遗传算法的样本集数据分割方法

提出了一种基于遗传算法的样本集数据分割方法。数据挖掘过程中该方法能够解决如何对一个样本集进行数据分割,从而得到最佳训练集和测试集的问题。通过该方法进行数据分割,不仅提高了分类模型的分类精度,而且能够最小化训练集和测试集之间的噪声百分比。最后,以一组软件项目样本数据为例说明该方法的有效性。     

基于决策树的遗传算法在数据挖掘领域的应用

论文详细阐述了基于决策树的改进的遗传算法的编码技术和相关遗传算子的操作;同时强调说明了相对于当前数据挖掘领域的数据分类算法,论文中的新分类方法在从海量数据库中全局优化搜索分类规则集所显示出它的优越性。最后,通过实例比较结果,证实论文中算法切实可行,有较高搜索效率。     

In-place random list permutations

We give two algorithms to randomly permute a linked list of length n in place using O(nlogn) time and O(logn) stack space in both the expected case and the worst case. The first algorithm uses well-known sequential random sampling, and the second uses inverted sequential random sampling.     

An improved representation for evolving programs

A representation has been developed that addresses some of the issues with other Genetic Program representations while maintaining their advantages. This combines the easy reproduction of the linear representation with the inheritable characteristics of the tree representation by using fixed-length blocks of genes representing single program statements. This means that each block of genes will always map to the same statement in the parent and child unless it is mutated, irrespective of changes to the surrounding blocks. This method is compared to the variable length gene blocks used by other representations with a clear improvement in the similarity between parent and child. In addition, a set of list evaluation and manipulation functions was evolved as an application of the new Genetic Program components. These functions have the common feature that they all need to be 100% correct to be useful. Traditional Genetic Programming problems have mainly been optimization or approximation problems. The list results are good but do highlight the problem of scalability in that more complex functions lead to a dramatic increase in the required evolution time.     

基于神经网络的智能过滤系统的研究与设计

首先对现有的网络安全隐患和现有的关于安全过滤方法作了简单介绍,提出现有网络过滤技术的不足,接着提出利用信息检索、人工智能和遗传算法相结合的技术,能有效提高网络过滤的性能.     

Enhancing the performance of hybrid genetic algorithms by differential improvement

A differential improvement modification to Hybrid Genetic Algorithms is proposed. The general idea is to perform more extensive improvement algorithms on higher quality solutions. Our proposed Differential Improvement (DI) approach is of rather general character. It can be implemented in many different ways. The paradigm remains invariant and can be easily applied to a wider class of optimization problems. Moreover, the DI framework can also be used within other Hybrid metaheuristics like Hybrid Scatter Search algorithms, Particle Swarm Optimization, or Bee Colony Optimization techniques.