CVPR18_TM_codes

SAR杂波统计建模方面的源程序，不仅可以能为SAR图像去噪工作提供数据源，更是研究SAR图像分割方面的很好的工具。(SAR clutter statistical modeling of the source, not only for SAR image denoising can provide the data source, but also aspects of SAR image segmentation a good tool.)

一个人体运动识别跟踪的示例程序，里面附带了一个运动视频，运行run.m文件。程序里包含了轮廓提取，跟踪算法，很好的tracking学习程序(A human motion recognition tracking sample program, which comes with a sports video, to run run.m file. The program includes a contour extraction and tracking algorithm, good tracking learning program)

利用DCT字典对图形进行稀疏表示，实现两幅图像的融合(Using the DCT dictionary to sparse representation of the graphics to achieve the fusion of two images)

织物疵点检测和识别系统.采用2中方法，有gui界面，支持实际纺织物的检测(Fabric defect detection and identification systems. Using 2 methods, gui interface, support for the detection of the actual textile)

c++语言或MATLAB编程对256×256的灰度图像进行8×8DCT变换，对变换后的结果进行处理： 采用区域编码和阈值编码方法实现图像压缩，保留 50％ 的系数(区域编码保留前 50%个系数，阈值编码保留 50%的大系数）； 再反变换后，对比原始图像和处理后的图像。(C++ or MATLAB programming performs 8 * 8DCT transform on 256 * 256 grayscale images, processing the transformed results: The region coding and threshold coding are used to compress the image, and the coefficient of 50% is retained (the first 50% coefficients are reserved for the region code and the large coefficient of the threshold code is retained by 50%). After inverse transformation, the original image and the processed image are compared.)

图像融合的质量评价指标，每个文件都有详细说明，每个指标都有相应的论文出处。亲测，可以使用。(Image quality uation of fusion, each file has a detailed description of each indicator has a corresponding paper source. Pro-test, can be used.)

多边形填充 多边形裁剪 自定义多边形区域 自定义裁剪区域(Polygon fill polygon clipping custom polygon area custom crop area)

实现基于样本块的图像修复，是Criminisi经典图像修复算法。(Criminisi algorithm about exemplar-based inpainting)

逆透视变换的算法,先生成一个图像 然后将图像坐标系和世界坐标系进行转换 最后将消除透视效果的图像显示出来(inverse perspective transformation algorithm Mr. images into one image will then coordinate system and the world coordinate system conversion for the final elimination of the effects of lens images show to)

用于图像分类的源代码，深度学习的应用，cnn卷积神经网络的使用，可以自己更换数据集或者调优，适合初学者使用！(For image classification of the source code, the application of in-depth learning, cnn convolutional neural network, you can replace the data set or tuning, suitable for beginners!)