卷积特征可视化

用平均背景法去除背景，算法思想比混合高斯模型和codebook模型简单，适合提取背景变化较小的场景。用中值滤波去除掉产生的椒盐噪声(Average method to remove the background with the background, the algorithm thought codebook than the Gaussian mixture model and a simple model for extraction of small changes in background scene. With a median filter to remove salt and pepper noise generated)

EM算法是最为经典的统计图像分割方法，代码基于matlab，使用EM算法实现图像分割,源代码加实例图像，非常好用(By using EM algrithm to realize the segementation of image)

Opencv处理图像，用ostu法实现图像的二值化(Opencv image processing, using ostu method to achieve image binarization)

基于opencv实时图像的2值化处理，一定要有摄像头(Opencv-based real-time image processing of two values, there must be a camera)

压缩感知中稀疏分解算法的正交匹配追踪算法OMP(An orthogonal matching pursuit algorithm for compressed sensing OMP)

小波算法。机械化的提高我们的效率。让我们减少错误率。大家要对她有性心。(Wavelet algorithm. Mechanization to improve our efficiency. Let us reduce the error rate. We have to have sex to her heart.)

说明：  方向梯度直方图检测算子，能够提取图像的HOG特征，进行分类与识别(Directional gradient histogram detection operator can extract the hog features of the image for classification and recognition)

图像检测的算法，Faster R-CNN算法，先对整张图像进行卷积计算，然后通过感兴趣区域池化层(RoI Pooling Layer)将选择性搜索算法推荐出来的候选区域和卷积网络计算出的特征映射图进行融合，得到候选区域对应的特征矢量，这种共享卷积计算的操作极大地减少了卷积计算的次数。而且这些特征矢量的维度统一，方便后续的分类工作。通过感兴趣区域池化层处理卷积特征，并将得到的特征送往两个并行计算任务进行训练，分类和定位回归。通过这些方法和改进的框架，Fast R-CNN 用更短的训练和测试时长，取得了比 R-CNN 更好的效果(Faster R-CNN algorithm first convolutes the whole image, then fuses the candidate regions recommended by the selective search algorithm and the feature mapping maps calculated by the convolution network through the RoI Pooling Layer to get the corresponding feature vectors of the candidate regions, which greatly reduces the number of convolution calculations. Moreover, the dimension of these feature vectors is unified, which facilitates the subsequent classification work. The convolution feature is processed by the pooling layer of the region of interest, and the obtained feature is sent to two parallel computing tasks for training, classification and positioning regression. Through these methods and improved framework, Fast R-CNN uses shorter training and testing time and achieves better results than R-CNN.)

本程序利用立方体进行标定，从而得到摄像机成像参数(This procedure used to calibrate the cube to get the camera imaging parameters)

实现了Chan－Vese算法，参考文献active contour without edges, 图像处理方面引用率最高的文献之一(Realize the Chan-Vese algorithm, reference active contour without edges, image processing, to quote one of the highest rates of the literature)