FFT

本程序可模拟ND情况下的傅里叶变换，经过验证无误(This program can simulate the Fourier transform of the ND situation, and after the test, the program is correct )

基于改进的广义S变换，能更好的对数据进行时频分析。(Modified S-transform and its application in data analysis)

用基于主元分析的方法对被控过程进行建模，可以消除变量之间的贡献性。(Accused of process modeling method based on principal component analysis, contribution between the variables can be eliminated.)

电容层析成像TSVD算法的演示实例以及详细的说明(An example of a demonstration of TSVD algorithm for capacitance tomography and a detailed description)

程序用Fortran95标准，给出了双共轭梯度法求解超大型稀疏矩阵的子程序。还给出了一个例子，例子的数据用索引存储了维数将近10万的对称稀疏正定矩阵。(Using Fortran95, a method of Biconjugate gradient for super linear sysem is provied. In addition, a example with the storage type of row-column index is also provide for test these subroutines. )

支持向量基预测的小数据量的预测方法。包括数据拟合，多项式预测等功能()

Hypercomplex mathematical morphology

计算改进型SVSLMS算法，在初始收敛速度，时变系统分析能力等方面优于传统的LMS算法(Improved SVSLMS calculation algorithm, the initial convergence rate, time-varying systems analysis capability is superior to the traditional LMS algorithm)

hartigans Sequential Leader Clustering Algorithm in terms of OpenCV (ver.1.1) Sequential Leader algorithm: Hartigan, J. A. (1975), Clustering Algorithms. John Wiley and Sons, Inc., New York, NY. 1. Select maximum cluster "radius" 2. Start with zero clusters 3. Add each item to be clustered to: * Closest cluster if distance <= r * New cluster if distance to closest cluster > r 4. Compute new center from members in cluster 5. Empty the clusters (keeping the centers) 6. Return to step 3 (until no changes are made)

说明：  基尔霍夫近似方法，计算目标的双站散射系数。参考乌拉比的微波遥感。(Kirchhoff s approximation method to calculate the bistatic scattering coefficient of the target. Refer to the Orabi microwave remote sensing.)