资源列表
DIC
- 通过识别一系列图像的变形得到物体的应力和应变。可以应用在试件试验中的非接触测量-Deformation of the identification of a series of images obtained by the stress and strain of the object. Non-contact measurement can be applied in the specimen test
GeoMatch_src
- 利用opencV进行基于形状的匹配，能够很好的实现在遮挡的情况下的匹配。-The use OPenCV shape-based matching, can be a good match in the case of occlusion.
DS-5_crack
- 使用期限到2099年 破解步骤参见crack.txt 只在5.11版本上破解过，其他版本未尝试。 如果破解后出现问题，卸载DS-5并删除C:\Users\Administrator\AppData\Roaming\ARM\DS-5（WIN7）文件夹，然后重新安装。 仅用于学习目的。-Use deadline to 2099 cracked steps, see crack.txt only 5.11 cracked version, other versi
nsf5
- 隐写方法nsf5 是f5的改进型 用matlab编写-steganalysis nsf5
2DDLPP
- 该源代码是人脸识别中的二维单项局部保持投影算法2DDLPP，源代码下载后就可以执行，简单，易理解。-The source code is a two-dimensional face recognition single locality preserving projection algorithm 2DDLPP, after downloading the source code can be executed, simple and easy to understand.
DAP_OSR
- 代码主要实现了零样本图像分类中的直接属性预测模型。在没有训练样本的情况下对其进行分类。-Code mainly realize the zero sample image classification of direct property prediction models. In case of the absence of training samples to classify them.
CNN-
- CNN卷积神经网络数字识别MATLAB代码-CNN convolutional neural network digital identification MATLAB code
libzbar-master
- 针对嵌入式开发而精简的ZBar，二维码识别代码-For embedded development, streamlined ZBar, two-dimensional code identification code
HS
- 基于光流法的人群异常行为识别，可以检测人群聚集和高速运动,zhu.m为主程序-Abnormal human behavior recognition based on optical flow method, can detect crowds gathered and high-speed movement, zhu. M program
Kinect实现PPT的手势翻页
- 使用Kinect在VS中可打开文件，然后可以进行PPT的手势翻页操作
c#图片文字识别
- c#基于Tesseract框架和语言训练库的图片文字识别。全工程文件，对仅含文字的清晰的图片，识别率较高！(C# text recognition based on Tesseract framework and language training library. The whole engineering document has a higher recognition rate for clear pictures containing only words!)
Car_100_5layers_English
- 车牌识别Python程序，使用五层的CNN网络，使用TensorFlow框架编程(car license plate recongition)