MatlabImageRetrieval

这是一重要的图像修补源代码，基于图像快的修补，是修补算法的经典制作（1）(inage quilting image inpainting )

人脸68点标注，用于人类面部关键点检测，结合dlib(face landmark master)

数字图像copy-move篡改检测。特征提取，相似块搜索(detection of copy-move forgery in digitial image)

这是个criminisi算法的源码，可以对图像进行相应的修补。效果很好。(This is a criminisi algorithm source code, images can be appropriate patch. With good results.)

本程序利用SIFT算法实现影像匹配的基础上，实现全景图的拼接，实用性很强，大家可以把拍摄的相邻的几张有重叠度的照片拼接成一张全景图。(The program uses SIFT image matching algorithm based on the realization of panorama stitching, very practical, we can put a few have taken the overlap of adjacent photos into a panoramic mosaic map.)

光斑中心提取算法，在VS2010下，用opencv2.3.1编写的控制台程序，主要功能是利用hession矩阵和高斯滤波进行光斑中心点提取，里面附测试图片，提取精度为亚像素级别。(Spot center extraction algorithm, in VS2010, written by opencv2.3.1 console program, the main function is to use hession matrices and Gaussian filter for extracting the center spot, which is attached test images, the extraction accuracy of sub-pixel level.)

运用于人脸图像预处理中，主要实现功能有：人脸检测，人脸对齐，人脸图像放缩，人脸剪切。(Applied to the face image preprocessing, the main features are: face detection, face alignment, face image zoom, face cuts.)

CNN中的手写识别LeNet5实现，用python，进行了中文注释(CNN handwritten recognition LeNet5 implementation, using python, carried out in Chinese Notes)

说明：  使用matlab搭建神经网络，从单目图像中估计深度值(Using matlab to build a neural network to estimate depth from monocular images)

该程序实现了harris角点的提取以及其在两幅影像上的配准结果，希望对大家有用(The program implements the harris corner extraction as well as their two images on the registration results, we hope to be useful)