calendar

凸组合滤波器CLMS，代价函数为e的平方，根据IEEE添加的限定条件，程序稳定，可运行，有详细注释(Convex combination of filters, the cost function is the square of the e qualification added by the IEEE, the program can run detailed notes)

一维大地电磁正则化反演。matlab的程序，用的是occam算法。(One dimensional magnetotelluric regularized inversion. Matlab program, using the Occam algorithm.)

时频分析工具包，可用于非平稳信号处理，for matlab(This is the Time-Frequency Toolbox, a collection of about 100 M-files , developed for the analysis of non-stationary signals using time-frequency distributions. It consists in groups of signal generation files, processing files and post-processing files (including visualization tools). The toolbox contains numerous algorithms for time-frequency signal analysis, with special emphasis on quadratic energy distributions within Cohen s class and the affine class, including most recent developments based on reassignment. The Time-Frequency Toolbox has been developed by Francois Auger, Olivier Lemoine, Paulo Gonlv and Patrick Flandrin during the years 1995-1996. Later, modifications for Matlab 5 were made available.)

为圆柱体定义一种有效地多边形网格表示，并调整对表示的选择 (《计算机图形学（第三版）》P139/3.25)(For the cylinder to define a polygon mesh that effectively, and that the choice of adjustment (" Computer Graphics (Third Edition)" P139/3.25))

该程序用于实现图像分割中的区域生长法。共有4个参数。1，2，种子点的座标，3，图像，4，生长时用到的域值。(this code is used for region growing in the image segmentation domain.)

非常适合计算机视觉方面的研究使用,仿真图是速度、距离、幅度三维图像，包括AHP，因子分析，回归分析，聚类分析。( Very suitable for the study using computer vision, FIG simulation speed, distance, amplitude three-dimensional image, Including AHP, factor analysis, regression analysis, cluster analysis.)

灰色系统的matlab实现程序!可以预测未来态势发展(Gray system procedures matlab realize! Can predict the future trend of development)

说明：  matlab编程引擎(matlab programming engine)

Schrodinger and Poisson solver

constructive role of noise in signal detection from parallel arrays of quantizers.(A noisy input signal is observed by means of a parallel array of one-bit threshold quantizers,in which all the quantizer outputs are added to produce the array output.s(t)=s1(t)with prior probability P0,or s(t)=s0(t)with prior probability P1=1-P0 input signal-noise mixture x(t)=s(t)+n(t) x(t) is observed by means of a parallel array of M threshold comparators. yi(t)=U[x(t)+wi(t)-theta],i=1,2...M.wi(t) is threshold noise.Y(t)=y1(t)+y2(t)+...yM(t), the array output Y(t) is meansured at T distinct times.the plot is fthe theoretical value of minimal detection error probability Per varied with sigma.)