不确定拓扑关系模糊推理

针对定性空间推理难以处理不确定空间拓扑关系推理的缺憾,将确定拓扑关系复合表推理方法扩展到模糊拓扑空间,建立了基于拓扑关系复合表的不确定拓扑关系模糊推理模型,设计了基于RCC5的拓扑关系复合表的模糊推理规则库和推理的方法流程,并结合具体计算实例对模糊推理模型进行了实验研究,推理结果验证了模型的认知合理性和可行性。     

基于ANN-FES的石漠化综合治理智能决策支持系统设计及应用研究

对人工智能技术的两个分支神经网络(ANN)与模糊专家系统(FES)各自的基本工作原理进行了阐述,分析了基于神经网络和模糊专家系统集成的混合系统功能框架,并介绍了系统结构及知识表示、知识获取、知识简化和推理机制等方面的基本方法.在此基础上,以都安石漠化综合治理智能决策为例,提出了基于神经网络和专家系统集成的石漠化智能决策系统的结构.人工神经网络实现石漠化预警分析,预测得到的石漠化危险性指数,最后得出专家系统所需的预警度(无警、轻警、中警和重警).模糊专家系统的推理机通过对神经网络得到的初步数据和其他测量的数据处理,实现系统的综合诊断,最后由治理模式系统确定采用哪种综合治理模式.该方法融合了神经网络自适应学习能力强和模糊专家系统知识表达明确的优点,简化了神经网络学习数据获取及模糊推理规则建立的过程.     

基于不精确推理的GES辅助决策系统研究

介绍了地理专家系统(GES)辅助决策系统的构建原理与实现方法,讨论了GES中专家知识的表示方法与推理控制策略。系统将领域专家知识划分成不同的层次和类型,采用框架网络数据结构和产生式规则表示地理学领域专家的专门性知识与经验性知识,使知识的表示简洁易于理解。在知识的推理策略中引入不精确推理机制,保证了推理机制的实现。以乌鲁木齐河流域为试验区,阐述了流域土地合理利用规划决策专家知识的表示方法,以及专家知识表示的基本规则和系统专家知识的组织方式。     

模糊目标的概念模型和应用

由于现实世界的连续性和多样性，地理实体具有不确定空间分布范围和边界，它们在GIS中的表达被称为模糊目标，提出了“模糊空间分布目标”的概念表示模糊目标即地理实体的复杂性和多样性。讨论了3个通用模糊目标模型以适用于不同的情形,FF－object具有模糊空间分布和模糊内核，适用于描述空间重叠的目标；FG－object具有确切内核和模糊空间分布，适用于描述空间分离但边界不明的目标，CF－object具有确切的条件边界和模糊内核，适用于描述空间分离且有边界的目标，我们用一个海岸带研究的实例说明模糊目标模型的应用。     

基于模糊本体的时空数据模型构建方法研究

针对时空推理及信息查询中存在的时空模糊性问题,通过模糊隶属函数对模糊本体中的模糊概念和关系进行描述,在模糊本体5W2H语义概念模型的基础上构建了FGO-5W2HST时空数据模型,研究了本体方法下的时空推理的模糊扩展,并将上述研究应用于旅游景点信息领域,从而对时空推理查询及语义信息检索过程进行模拟验证。     

用专家系统技术确定地图要素的表示等级

本文在总结了地图要素分类分级知识的基础上，提出了地图要素表示等级的概念，用于描述事物的概括和抽象，同时分析了影响地图要素表示等级的因素，介绍了知识的表示和推理，以及系统的功能。     

基于模糊本体的时空推理查询方法研究

针对时空推理及信息查询中存在的时空模糊性问题,通过模糊隶属函数对模糊本体中的关系进行描述,在模糊本体5W2H语义概念模型的基础上,研究了模糊本体下的语义处理方法和推理机制,基于本体模型和规则模型探讨了模糊本体知识推理机的设计方法,并将上述研究应用于旅游景点信息领域,从而对时空推理查询及语义信息检索过程进行模拟验证。     

三维体目标间拓扑关系与方向关系的混合推理

重点研究了三维空间中拓扑关系和方向关系间的混合空间关系推理。用Allen区间关系对描述基于投影的空间划分方法得到的方向区域和用九交矩阵描述的拓扑关系,用定义法研究混合空间关系推理,推理结果用组合推理表表示。     

矿岩界线的自动推连

着重讨论了如何利用人工智能技术进行矿岩界线的自动推断和连接，包括对所研制的基于知识的矿岩界线自动推连系统的描述；关于数据与知识的表示和组织、推理过程、策略以及系统实现等内容。     

模糊分析学在空间信息分析中的应用初探

本文综述了空间现象分布和空间信息分析中的模糊问题，给出了模糊区域的概念及其测度方法，结合实例阐述了不精确空间关系分析，不精确空间关系综合及模糊推理在空间信息分析中的应用方法。     

自适应模糊规则分类方法及在TM土地覆盖分类中的应用研究

根据自组织网络和模糊逻辑推理,实现土地覆盖自适应模糊规则分类方法。该方法通过网络的节点和权值提取出模糊规则,调整网络中节点个数(即相应增加规则节点数)和权值向量,使模糊规则自动生成,并利用模糊逻辑推理,完成TM土地覆盖分类。对拒分类的像元,自适应增加K值使其可分。该方法所得分类精度及Kapp系数与最大似然分类方法结果相比分别提高了2.7%和2.9%;与自组织网络相比,总精度相差不大,而Kapp系数低1%。实验证明,如何提取和表示非光谱知识,从而解决类别混淆等问题,是提高自适应模糊规则分类性能的关键     

Fuzzy Neural Network Models for Supervised Classification: Multispectral Image Analysis

Abstract

It has been well established that neural networks provide a reasonable and powerful alternative to conventional classifiers. During the past few years there has been a large and energetic upswing in research efforts aimed at synthesizing fuzzy logic with neural networks. This combination of fuzzy logic and neural networks seems natural because two approaches generally attack the design of “intelligent” systems from quite different angles. Neural networks provide algorithms for learning, classification, and optimization whereas fuzzy logic deals with issues such as reasoning on a higher (semantic or linguistic) level. Consequently the two technologies complement each other. In this paper we propose two novel fuzzy‐neural network models for supervised learning. The first model consists of three layers, and the second model consists of four layers. In both models, the first two layers implement fuzzy membership functions and the remaining layers implement the inference engine. Both models use the gradient decent technique for learning. As an illustration, we have analyzed two Thematic mapper images using these models. Results are presented in the paper.     

Automatic Surface Reconstruction Based on the Fusion of Fuzzy Logic and Robust Estimation Techniques

Automatic surface reconstruction by means of digital image matching essentially involves computation in both image and object spaces. In image space, image matching is performed according to certain radiometric and geometric similarity criteria for conjugate point determination. The matching operations may therefore be regarded as decision–making processes. In practice, however, the similarity criteria for conjugate point determination demonstrate non–deterministic behaviour. Thus, the nature of these decision–making processes is more compatible with fuzzy reasoning concepts. In object space, on the other hand, surface reconstruction procedures, which operate on the basis of previously determined conjugate points, are essentially deterministic in nature and hence demand rigorous geometric modelling. In this paper, a new approach for automatic surface modelling is proposed, based on two different mathematical treatments: a fuzzy logic reasoning method for the conjugate point determination in image space; and a robust finite elements approach for the surface modelling in object space. Tests carried out on real data demonstrate the high potential of the strategy proposed.     

A fuzzy logic map matching for wheelchair navigation

Global Positioning System (GPS) plays a central role in solving the location-based needs of wheelchair users and efficient route guidance. An essential process in wheelchair navigation is matching the position obtained from GPS or other sensors on a sidewalk network. This process of map matching in turn assists in making decisions under uncertainty. However, GPS-based wheelchair navigation systems have difficulties in tracking wheelchairs in urban areas due to poor satellite availability. To overcome this, a fuzzy logic-based algorithm is applied to effectively perform matching wheelchair movements on sidewalks. Fuzzy logic, based on fuzzy reasoning concepts, in many circumstances can take noisy and imprecise input to yield numerically accurate output. Experiments on a university campus validate it practicable.     

Classifying Human Activity Patterns from Smartphone Collected GPS data: A Fuzzy Classification and Aggregation Approach

Smartphones have emerged as a promising type of equipment for monitoring human activities in environmental health studies. However, degraded location accuracy and inconsistency of smartphone‐measured GPS data have limited its effectiveness for classifying human activity patterns. This study proposes a fuzzy classification scheme for differentiating human activity patterns from smartphone‐collected GPS data. Specifically, a fuzzy logic reasoning was adopted to overcome the influence of location uncertainty by estimating the probability of different activity types for single GPS points. Based on that approach, a segment aggregation method was developed to infer activity patterns, while adjusting for uncertainties of point attributes. Validations of the proposed methods were carried out based on a convenient sample of three subjects with different types of smartphones. The results indicate desirable accuracy (e.g. up to 96% in activity identification) using of this method. Two examples are provided in the Appendix to illustrate how the proposed methods could be applied in environmental health studies. Researchers could tailor this scheme to fit a variety of research topics.     

基于Jena和地理本体的空间查询与推理研究

Jena根据OWL本体定义将地理数据进行RDF资源标注,形成带有语义信息的数据,便于地理空间数据的语义查询和推理。深入讨论Jena在基于地理本体的空间信息查询和推理中的应用机制,将用户针对数据源的查询转换为对地理本体的查询;利用建立在逻辑基础上和空间关系基础上的地理本体,发现空间数据源中一些隐含的概念或关系。结合实例,实现基于本体的地理空间信息查询和在自定义规则下利用Jena推理机的空间推理。     

八方向模糊不均匀划分及参考点位误差所致不确定性分析

方向概念的模糊性和空间数据固有的不确定性导致了方向关系的复杂性,在空间关系描述和推理研究中需要考虑空间数据的不确定性和方向概念的模糊性。在四方向模型中各方向片是等角划分;在八方向模型中4个主要方向片各占60°,4个次要方向片各占30°。利用区间二型模糊集理论建立了顾及参考点点位误差的八方向模糊不均匀划分模型,基于区间二型模糊集讨论了方向主隶属度成员函数和隶属度的不确定性。对比分析了八方向模糊不均匀划分模型与锥形模型的区别,讨论了具有点位误差的参考点与线和多边形的方向关系计算过程,通过两个实例分析了该模型的特点和点与多边形方向关系的确定方法。     

耕地需求量预测的加权模糊-马尔可夫链模型

以耕地利用动态度为度量，利用模糊有序聚类方法将耕地需求量划分为不同的模糊状态区间，利用模糊集理论构建马尔可夫链状态转移概率矩阵，以规范化的各阶自相关系数为权重改进传统的马尔可夫链预测模型，用改进后的模型对土地利用规划中耕地需求量进行预测。实验结果表明，改进后的方法较传统的预测方法更具科学性和实用性。