基于模糊神经网络的模糊空间对象生成方法研究  

Research on the Generation Method of Fuzzy Spatial Objects Based on Fuzzy Neural Networks

作　　者：张辉[1] 杨俊涛[2] 李参海 

机构地区：[1]中国矿业大学环境与测绘学院,江苏徐州221008 [2]中国测绘科学研究院重点实验室,北京100039 [3]黑龙江第三测绘工程院,黑龙江哈尔滨150086

出　　处：《测绘与空间地理信息》2008年第1期1-3,共3页Geomatics & Spatial Information Technology

基　　金：国家自然科学基金(40571127)资助

摘　　要：随着对GIS中的空间对象模型和自然地理特征表达的研究深入,模糊空间对象被提出。针对模糊空间对象表达的特点,提出了一种基于模糊神经网络的模糊空间对象生成方法。该方法将模糊技术与神经网络相结合,利用神经网络的学习能力调整模糊隶属函数和模糊规则,使系统具备自适应的特性。实验表明,这种基于模糊神经网络的生成模糊空间对象的方法比传统方法大大的提高了成果的精度。With the deep research on the spatial objects model of GIS and the representation of natural geographical feature, this paper put forward the fuzzy spatial objects. Referring to the characteristics of the representation of fuzzy spatial objects, it also brought forward a generation method of fuzzy spatial objects based on fuzzy neural networks. By combining the fuzzy technology and neural networks, utilizing the learning ability to enhance the fuzzy Membership Function and fuzzy rules, the system will be self - adaptive. From the experiments of this paper, by comparing with the traditional fuzzy objects generation, the method in this paper improved the accuracy of results.

关 键 词：模糊空间对象 模糊神经网络 模糊土地覆盖 

分 类 号：TP311.134[自动化与计算机技术—计算机软件与理论]