基于典型相关性的线性鉴别分析，先用PCA对数据降维之后，再结合典型相性鉴别分析来得到转换矩阵-Based on a typical linear correlation analysis to identify, first PCA for data reduction, the combined analysis to identify the typical phase of the transformation matrix obtained

Image zooming using directional cubic convolution interpolation