CAJ | 학술논문

为了获得更多的信息,越来越多的数据利用多路传感器进行采集,由此产生了大量的超高维时间序列.特征的提取在处理和传输这些数据中起到至关重要的作用.为此,提出一种最优鉴别平面(ODP)技术以消除数据冗余.该平面由两个在Fisher准则基础上建立起来的相互垂直的矢量组成,将模式样本投影到ODP上可得到二维特征矢量.为了衡量特征的有效性,分别用二次判别函数分类器和阈值矢量分类器对特征进行分类测试.同时,以心电信号为例对ODP方法进行测试,结果表明,该方法应用于超高维数据的特征提取是行之有效的.
위료획득경다적신식,월래월다적수거이용다로전감기진행채집,유차산생료대량적초고유시간서렬.특정적제취재처리화전수저사수거중기도지관중요적작용.위차,제출일충최우감별평면(ODP)기술이소제수거용여.해평면유량개재Fisher준칙기출상건립기래적상호수직적시량조성,장모식양본투영도ODP상가득도이유특정시량.위료형량특정적유효성,분별용이차판별함수분류기화역치시량분류기대특정진행분류측시.동시,이심전신호위례대ODP방법진행측시,결과표명,해방법응용우초고유수거적특정제취시행지유효적.
The increasing hyperdimensional data are acquired from multichannel sensors due to contain more significant information,but the amount of data becomes very large.Extracting significant features from these data is essential for processing and transmission.Optimal discrimination plane (ODP) technique was developed to reduce the redundant data.The features were extracted from the data using the ODP based on Fisher's criterion method.The patterns were projected onto the two orthogonal vectors that built up the ODP,and two-dimensional feature vectors were attained and utilized as features to represent the patterns.Electrocardiogram signals are applied to the analysis as an example in this study.A quadratic discriminant function based classifier and a threshold vector based classifier were employed to measure the performance of the extracted features,respectively.The results show that the proposed ODP is an effective and feasible technique to extract the features from the hyperdimensional time series data.