摘要
针对传统虚拟仪器数据库识别方法无法实现多变量空间尺度的识别,识别效果不佳的问题,提出对虚拟仪器数据库多变量进行空间尺度识别的方法。对虚拟仪器数据库中的多变量进行提取,并采用ICA方法和小波滤波器对提取的多变量数据进行预处理,得到数据的空间尺度信息;将多变量的空间尺度信息视为变量固有属性,根据变量固有属性形成概念分层,依据概念分层中的相关概念对多变量进行空间尺度划分;采用小波变换对多变量划分的结果进行重构,得出多变量空间尺度识别结果。通过多次仿真证明,所提方法可高效地实现对多变量空间尺度的识别,且识别效果较好。
Traditional virtual instrument database recognition method cannot achieve the multi-variable spatial scale recognition.Therefore,a method to recognize spatial scale of multiple variables in virtual instrument database was presented.Firstly,multiple variables in virtual instrument database were extracted,and then the extracted multi-variable data were preprocessed by ICA method and wavelet filter.After that,the spatial scale information of data was obtained.The multivariable spatial scale information was regarded as the inherent attribute of variable.On this basis,the concept hierarchy was formed and the spatial scale of multiple variables was classified by the related concepts in conceptual hierarchy.Finally,the wavelet transform was used to reconstruct the results of multivariable division,thus the result of multivariable spatial scale recognition was obtained.Simulation results show that the proposed method can efficiently achieve the recognition of multi-variable spatial scale.
作者
徐虹
彭雅丽
XU Hong;PENG Ya-li(Department of Software,Jiangxi Normal University,Nanchang Jiangxi 330022,China)
出处
《计算机仿真》
北大核心
2020年第1期435-438,共4页
Computer Simulation
基金
江西省高等学校科学研究项目(NJZY18207)。
关键词
虚拟仪器
数据库多变量
空间尺度特征
识别
Virtual instrument
Database multivariate
Spatial scale feature
Recognition