基于主元分析的频谱整体识别方法-JournalofNortheasternUniversity.PDFVIP

第29 卷第9期 东 北 大 学 学 报 ( 自 然 科 学 版 ) Vol29, No. 9 2 00 8 年 9 月 Journal of Northeastern U niversit ( Natural Science) Sep. 2 0 0 8 1 1, 2 1 1 李允公 , 张金萍 , 吴宁祥 , 刘 杰 ( 1. , 110004; 2. , 110142) : , PCA N N , , P CA , , , , , : ; ; ; ; : TP 273 : A : 1005-3026( 2008) 09-1322-04 A PCA-Based Integrative Spectrum Identification Method 1 1, 2 1 1 L I Yun-gong , ZH A N G Jin-p ing , W U Ning-x iang , LI UJie ( 1. School of Mechanical Engineering Automat ion, Northeastern Universit , Shen ang 110004, China; 2. School of Mechanical Engineering, Shen ang Institute of Chemical T echnolog , Shen ang 110 142, China. Correspondent: LI Yun-gong , E-mail: gli @ mail. neu. edu. cn) Abstract: Reducing the computational complex it is indispensable for the pattern identification in according to the integrative spectrum data. A PCA-based integrative spectrum identification method is therefore proposed. It regards an N-point spectrum as a point in the N-dimension space and forms a data matrix b use of the know n spectrum as samples. Then, after PCA, the number of directions of principal components satisf ing the threshold values of information remaining to reduce the dimensions of high dimensional data. And the centers of projective points of various spectrum in low dimensional space are computed to obtain the corresponding data templates. In applications, the identification results are classified according to the criterion w hich implies the shortest distance. The results of numerical simulation and voice recognition reveal that