hungarian-rvm

首先利用形态学开运算将图像中小于结构元素的亮点去掉，利用源图像与处理后的图像作差，对差图像进行阈值分割，得到二值图像，将二值图像中的一些干扰区域去掉，提取二值图像中的圆形区域，并在源图像上对小目标进行标记，实现对圆形亮目标的检测(Remove the area in the image smaller than structural element by morphologic open operation，receive image A，original image subtract image A，threshold segmentation for difference image，receive binary image，remove the interference region in the binary image，then extract circular area，and sign the target in the original image. Realize the circular bright target detection)

matlab程序添加水印的DCT算法，先将水印加密，然后嵌入图像中(matlab program to add a watermark in the DCT algorithm, first the watermark encryption, and embedded images)

cv阈值去噪方法和贝叶斯阈值去噪方法，这两种方法是很常见的阈值去噪方法。(cv thresholding method and the Bayesian thresholding methods, the two approaches are very common thresholding methods.)

进行kcm和fcm聚类分割图象的性能比较(kcm for clustering segmentation and image fcm Performance Comparison)

多幅图像拼接matlab实现 sift特征提取、描述、匹配、RANSAC、仿射变换(Multiple image mosaics, MATLAB, SIFT feature extraction, description, matching, RANSAC, affine transformation)

对于MATLAB双目视觉处理中，通过前期获取摄像机参数等后制成深度图。(For MATLAB binocular vision processing, we can get the depth map after obtaining camera parameters in the early stage.)

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.)

说明：  一种基于tv模型的彩色图像修复方法，效果很好(Tv model based on color image restoration)

使用Tensoflow实现图卷积神经网络模型(Use Tensoflow Implementation Figure convolution neural network model)

说明：  一个SAR目标边缘自动检测与定位算法的一个车辆检测的视频(A vehicle detection, an automatic detection and localization of SAR edge algorithm of video)