image

该VC++程序是基于CUDA架构的中值滤波程序，GPU加速，并且使用了opencv函数库(The VC++ program is based on the CUDA architecture median filtering program, GPU acceleration, and use the opencv library)

内含完整基于图像处理的缺陷检测源代码，可以直接运行(Including a complete defect detection based on image processing source code can be directly run)

用SVD算法实现图像滤波去噪处理，提高图像的信噪比。(SVD algorithm using image processing filter to improve image quality.)

压缩文件是CNN卷积神经网络图像处理的数据库mnist_uint8.mat(Compressed file is the database of CNN convolution neural network image processing mnist_uint8.mat)

通过计算机usb摄像头进行图像采集，然后实时处理，跟踪其中运动物体，并进行计数，采用中值滤波法得到背景，并根据需要进行更新，视频处理采用了动态差分与前景差分结合的方法，效果比较理想，但无法解决人物重叠的问题(Usb camera through the computer for image acquisition, and real-time processing and tracking moving objects in which, and the count median filter by using the background and update as needed, video processing using a dynamic difference with the prospect of differential combination of methods, results ideal, but can not solve the problem of overlapping characters)

说明：  短时傅里叶变换的一些简单的演示，其中有矩形窗和高斯窗(some demos for the short time fourier transform)

对比度拉伸，对原始图像进行对比度拉伸，从而得到拉伸后的图像(tensile contrast, the original image contrast stretching, and thus drawing the images)

在张正友的基于平面方格的摄象机标定中,单应矩阵是关键的一步,本程序是效率精度比较高的源码(Zhang is the Friends of the plane on the box camera calibration alone should matrix is a crucial step, The procedure is relatively high efficiency accuracy of source)

为GABOR小波的详细MATLAB实现代码，详细分析了该小波的各种性能()

说明：  通过matlab源码实现图像的最佳自动阈值分割，所选算法是常见的大津法(Matlab source code through the realization of the best automatic image thresholding, the selected algorithm is a common method of Otsu)