psnr

N年前上数字图像课时写的实验报告和程序，用滤波反投影（Filter Back Project，FBP)法重建图像。(Report and source code for data image restore experiment, using FBP.)

数字图像快速傅里叶变换（FFT）。将图像转化成二维灰度图像后再做FFT变换，并输出变换后的幅值图。(Digital image fast Fourier transform (FFT).The image is transformed into a two-dimensional gray level image and then the FFT transformation is made, and the amplitude diagram is output after the transformation.)

相变凝结蒸发，源项vof-udf，定义连续性方程，能量方程，动量方程中的质量源项，能量源项和动量源项(condensation and evaporation)

说明：  尼康相机拍摄的原始数据RAW格式，matlab实现该类格式的读取，显示。便于计算机进行后续的处理。直接运行即可。(RaW format of raw data captured by Nikon camera, which is read and displayed by matlab. It is convenient for computer to carry out follow-up processing. It can run directly.)

gongzhenQ.m:计算双稳态随机共振系统输入输出信信噪比增益，非常明显的随机共振现象。，已通过测试。 (gongzhenQ.m: calculation of the stochastic resonance of bistable stochastic resonance system input output signal SNR gain, it is clear. , Has been tested.)

由于图像在经过一次 JPEG 压缩和双重 JPEG 压缩之后其 DCT 系数具有一定的相关性，通过对这种相关性进行分析，提出了一种基于 JPEG 图像双重压缩特性的图像篡改检测算法，并用此算法检测图像是否经过篡改和定位篡改区域。 (Since the image after a first JPEG compression JPEG compression and double its DCT coefficients having some relevance, by this correlation is analyzed, and an image based on JPEG image compression characteristic double tamper detection algorithm, and the algorithm used to detect whether the image been tampered with and locate the tampered areas.)

经典数字图像处理算法仿真，包括图像的傅里叶滤波及压缩，图像的DCT高通、低通滤波，图像直方图均衡化，图像平滑与锐化，图像的模糊化，哈夫曼编码等，以及GUI图形化界面。(Classic digital image processing algorithm simulation, including Fourier filtering and image compression, image DCT high-pass, low-pass filtering, image histogram equalization, image smoothing and sharpening, blurring the image, Huffman coding, as well as GUI graphical interface.)

harris, harris-laplace方法，特征点检测方法总结(harris, harris-laplace method, Feature Point Detection Methods)

根据RBM深度网络（Hinton，2006）进行MINST手写文字的分类器的训练，利用训练得到的权值制作了这个小程序，通过这个程序可以看出训练结果对数据集内的测试样本和训练样本都能进行很好的识别，但是对其他的手写字体识别就没有那么好了。(According to RBM depth network (Hinton, 2006) conducted MINST handwritten text classifier training, using the training to get the right value to produce this small program, this program can be seen through the results of the training data set and the training sample test samples are can be well identified, but on the other handwriting recognition is not so good.)

最完整的CCA典型相关分析程序，分析两幅图像，相似度，匹配，融合都可以使用，分析天气变量也可以使用。(The most complete CCA canonical correlation analysis program, analysis of two images, similarity, matching, fusion can be used, analysis of weather variables can also be used.)