The small sample size problem is often encountered in pattern recognition. It results in the singularity of the within-class scatter matrix S,? in Linear Discriminant Analysis (LDA). Dijjrent methods have been proposed la solve this problem in face

估计类均值向量和协方差矩阵的方法。 类间离散度矩阵、类内离散度矩阵的计算方法 -Estimated class mean vector and covariance matrix method. Between-class scatter matrix, within-class dispersion matrix calculation method

matlab经典绘图案例。包括三维曲面z=x^2+y^2的动画、正弦曲线的动画、切片图、向量图、三维彗星图、三角形网格图、三角形表面图、球面、等值线图、三维瀑布图、三维火柴杆图、三维散点图、柱形图、三维条形图、三维饼图、带形图、曲线图、高斯分布函数的曲面图、高斯分布函数的网格图、5阶高斯分布矩阵、梯度场矢量图、罗盘图、玫瑰花图、羽列图、数据向量的单轴对数坐标图、矩阵的条形图 矩阵的面积图、二维饼图、散点图、误差图、火柴杆图和碗状曲面图等等。 -Drawing matlab classic

绘制二维密度散点图，根据x，y的最大最小值将坐标分成若干小格，然后统计落在每个小格里的点数，最后形成的就是一个矩阵，然后直接出图-Drawing two-dimensional density scatter plot, according to the x, y minimum and maximum values ​ ​ of the coordinates into a number of small cells, then every little statistic