zftjh

高斯粗糙表面的散射系数，输入参数含波长，需要regeng.m产生高斯粗糙表面。(rs1dg computes the bistatic scattering coefficient for Gaussian rough surfaces with Gaussian spectrum.)

文件为图形的显示函数，如单个像素的显示，直线的显示，字符的显示(File for the graphic display function, such as a single-pixel display, a straight line display, character display)

一个不错的人脸检测演示系统里面有人脸检测的实际照片包+(face detect system)

基于不精确拉格郞日算子法的低秩矩阵重构程序，可以用于图像分割，将目标图像分割为背景和前景，从而将前景分离出来。(Lagrangian method based on imprecise low rank matrix reconstruction procedures can be used for image segmentation, the target image into background and foreground, which will separate the foreground.)

说明：  Calculate the Laplacian of an image, can be accelerated by the GPU.

通过使用块匹配法实现运动矢量的检测，块大小和搜索区域大小可自定(Achieved through the use of block matching motion vector detection, block size and search area size can be customized)

1．用Matlab产生正弦波,矩形波,以及白噪声信号，并显示各自时域波形图 2．进行FFT变换，显示各自频谱图，其中采样率，频率、数据长度自选 3．做出上述三种信号的均方根图谱,功率图谱,以及对数均方根图谱 4．用IFFT傅立叶反变换恢复信号，并显示恢复的正弦信号时域波形图( 1. Using Matlab generated sine wave, square wave, and the white noise signal, and displays the respective time-domain waveform 2. The FFT transform, showing each spectrum, where the sampling rate, frequency, data length choice 3. Rms signal to make the three maps, power spectra, as well as the number of rms spectra 4. With the inverse Fourier transform IFFT restore the signal and displays the recovered time-domain waveform sinusoidal signal)

图像处理中的基本问题 做对比的试验(Image processing compared to the basic problems of the pilot)

令x(n)=5exp(j*0.15*n*n)+6exp(j*300n-j*0.15*n*n) ，w(n)为高斯窗函数。试用matlab软件，取不同长度的窗函数，分别求x(n)离散短时傅里叶变换，并进行信号重构。(ORDER x (n) = 5exp (j* 0.15* n* n)+6exp (j* 300n-j* 0.15* n* n), w (n) for the Gaussian window function. Trial Matlab software, take different lengths of the window function x (n), respectively, seek to discrete short-time Fourier transform, and to reconstruct the signal.)

CNN中的手写识别LeNet5实现，用python，进行了中文注释(CNN handwritten recognition LeNet5 implementation, using python, carried out in Chinese Notes)