摘要
主分量分析(PCA)是统计学中分析数据的一种有效的方法,可以将数据从高维数据空间变换到低维特征空间,因而可以用于数据的特征提取及压缩等方面。在该文的形状识别系统中,用PCA法提取图像的形状特征,能够较好地满足识别层的输入要求。在识别层研究了3种识别方法:最近邻法则、BP网络及协同神经网络方法,均取得了满意的实验效果。
The principal components analysis(PCA)method,which is an effective method of analyzing data in statistics,is widely used in the data feature extraction and data compression for the higher dimensional data space can be transformed into the lower dimensional feature space by the PCA method. In this paper, the image shape feature extracted by the PCA method can meet the needs of the recognition layer well. In the recognition layer,three methods, i.e. the nearest neighborhood algorithm,the BP neural network method and the synergetic neural network method, are discussed. The recognition results show that all the three methods are effective.
出处
《合肥工业大学学报（自然科学版）》
CAS
CSCD
2003年第2期176-179,共4页
Journal of Hefei University of Technology：Natural Science
基金
国家自然科学基金资助项目(60175011)
安徽省自然科学基金资助项目(01042301)
教育部优秀青年教师计划项目
