fast-motion--deblurring

针对齿轮滚动轴承等的早期损伤类故障, 提出将小波包分解作为包络分析的前置处理手段以提取振动信号的故障信息特征 。 在简述小波包基本原理的基础上, 通过仿真信号, 对振动信号的具体处理过程进行分析, 并对可能遇到的问题, 提出处理办法, 然后应用于诊断实例 。 (Early damage fault for rolling bearings and other gear, the proposed wavelet packet decomposition as pre-processing means envelopment analysis to extract characteristic vibration signal fault information. On the basis of the basic principles outlined wavelet packet, through simulation signal processing for specific vibration signals are analyzed, and the problems that may be ncountered, proposed approach, and then applied to the diagnosis example.)

说明：  血管骨架提取文件。冠脉血管树重建冠状动脉造影图像的三维重建是一个复杂的过程 主要包括血管骨架提取与表达、血管段匹配、骨架点匹配、三维骨架重建、表面重建等步骤。(Extraction file of vascular skeleton. Three dimensional reconstruction of coronary angiography image is a complex process, which mainly includes the extraction and expression of vascular skeleton, vascular segment matching, skeleton point matching, 3D skeleton reconstruction, surface reconstruction and so on.)

说明：  基于W-SVD的小波数字水印算法，包括水印生成、水印嵌入和水印检测，理解SVD奇异值分解算法。(Wavelet digital watermarking algorithm based on W-SVD includes watermarking generation, watermarking embedding and watermarking detection, and understanding SVD singular value decomposition algorithm.)

说明：  非下采样contourlet变换是一种较新的多尺度几何分析方法，能够很好地描述信号的线奇异特征，具有多方向滤波器组。(Nonsubsampled contourlet transform is a relatively new method of multiscale geometric analysis, can well describe the signal characteristics of the line singularity, with multi-directional filter.)

检测视频中的运动目标以及阴影，并消除阴影，基于opencv，将RGB颜色空间根据一定的转换关系转换到HSV空间，最后消除阴影， 可以在vc6.0和vs2008上正常运行(Shadow detection algorithm based on HSV color space)

本程序采用了特征线法计算管道阀门关闭时的水锤(This program uses a water hammer characteristic line method to calculate the pipe when the valve is closed)

利用卷积神经网络进行特征提取，并将卷积特征可视化(Feature extraction using convolutional neural networks and visualization of convolution features)

一个C#使用WebService调用海康SDK实现抓图与录像实时下载的小例子

利用图像的不变矩以及模式识别的内容，构造三维飞机图像的矩特征，然后再利用D-S证据理论的改进方法 吸收法对图像进行融合 对飞机的型号进行匹配(Moment invariant of image and pattern recognition of the content, structure features three-dimensional plane images of the moment, and then use the improved DS evidence theory method of absorption of the fusion images to match the type of aircraft)

将图片通过灰度值转化为文字符号，并排列为原来图像的样式。 (Convert image by gray scale value into text symbols, side by side as the style of the original image. )