fenjiechonggou

该类计算图像的主分量,特征值,特征向量,并且使用主分量重构.(The main components of such calculation of the image, feature values, feature vectors, and the use of principal component reconstruction.)

图像二值化算法，试用效果还不错，欢迎大家试用(Improved Image and Text Binarization Algorithm)

指纹图像处理包括梯度处理、方向场处理、平滑处理、智能增强处理等(The fingerprint image processing including gradient processing, direction field processing, smooth processing and intelligent enhancement processing, etc)

利用侧抑制原理编写的一个最基本的小程序。程序实现了初步的轮廓提取效果。可供想要入门侧抑制原理的朋友参考。(The use of lateral inhibition, a basic principle of writing a small program. Program to achieve the initial contour extraction results. Introduction to principles of lateral inhibition for want of a friend for reference.)

对计算图像的LBP矩阵并具备旋转不变性，用于目标检测和识别(The LBP matrix of the calculated image has the rotation invariance for target detection and recognition)

评价图像边界的好坏，FOM值越趋于1质量越好。(figure of merit to objectively evaluate the performance of edge detectors)

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.)

摄影测量单独法相对定向源代码的编写，结构较简单，易懂(relative oriention)

光流法实时实现的文献及代码，2012发表于cvpr，200*200的图像，每秒可达200帧，2012_CVPR_Lip-motion events analysis and lip segmentation using optical flow(Optical flow method to achieve real-time literature and code, 2012 published in cvpr, 200* 200 images per second, up to 200, 2012_CVPR_Lip-motion events analysis and lip segmentation using optical flow)

完整的图像处理程序，完成基本的图像处理功能，如，各种二值化算法的实现，图像边缘检测，图像分割，降噪等。其中有车牌识别实例。(Complete image processing program, the completion of the basic image processing functions, such as, a variety of value of implementations of the algorithm, image edge detection, image segmentation, noise reduction. License plate recognition instance.)