HazeRemoval

图像拼接，用于合成全景图像，VC++开发环境，C++实现。(Image Stitching)

本程序实现了对影像特征点自动提取，利用Morevac、Forstner、Harris3个经典算子。在此基础上利用相关系数法实现影像自动匹配，并且引入最小二乘平差，使匹配点精度有所提高。 在搜索点过程中，利用了核线影像特性，对二维影像搜索使用了爬山法启发式搜索。对大数据量影像采用影像金字塔结构处理。 1、使用GDAL库读取影片，支持TIFF、PNG、JPEG、JPG、BMP、GIF、IMG格式读取。使用GDI绘图。 2、防止大数据量绘图视图闪烁，图片显示采用双缓存技术。 3、保存视图数据为图片文件，支持TIFF、PNG、JPEG、JPG、BMP、GIF格式保存。 4、TreeCtrl控件、ListCtrl控件的基本操作。 5、MFC单文档程序视图通讯、更换视图、视图分割。(This program of image features automatic extraction point, use Morevac, Forstner, Harris3 a classic operator. Based on the correlation coefficient method using the image matching, realization and introducing least squares adjustment, the matching point to improve precision. In the search process, the use of the image characteristics of nuclear line, 2d image search using the heuristic method of climbing. For large quantity of data using image processing image pyramid structure. 1, using film, read GDAL repository support JPEG, TIFF, PNG, GIF, BMP, JPG, read IMG format. Use GDI drawing. 2 and prevent large amount of data, drawing view pictures showed adopt double caching techniques. 3, save for the image file, views data support JPEG, TIFF, JPG, PNG, GIF, BMP format. 4 and TreeCtrl controls, the basic operation ListCtrl controls. 5 and MFC single document procedure view communication, replace view, view.)

能够计算傅里叶描述子，并对图像进行重构，实现对目标图像和源图像的辨认(Fourier descriptors can be calculated, and the image reconstruction, image and the source image to target identification)

包括基于SIFTGPU的sift特征提取，sift匹配，基于ransac误匹配点剔除等(SIFTGPU based sift feature extraction, sift matching, mismatching points based ransac removed, etc.)

一个很实用图象处理原码(a practical source image processing)

程序可以实现BMP图片特效处理,包括浮雕效果(bmp picture)

1.打开左右两张影像2.Harris算法提取同名点3.显示提取同名点（若要进行下一步匹配则不建议显示，这里只是用于查看是否提取出）4.相关系数匹配（只在找出的同名点内匹配）5.LSM最小二乘匹配（先进行粗匹配，再精匹配）(1 Open the left and right images 2.Harris algorithm to extract two tie points 3 shows the extraction of corresponding points (not recommended to proceed to the next match show, here is used to see if extracted) 4. Correlation coefficient matching (just looking match within a point of the same name) 5.LSM least squares matching (rough first match, then fine match))

根据D. Comaniciu, P. Meer: Mean Shift: A robust approach toward feature space analysis 以及 C. Christoudias, B. Georgescu, P. Meer: Synergism in low level vision.这两篇文献提供的方法编写的图像分割代码，作者是 Chris M. Christoudias, Bogdan Georgescu，代码经我看了后加了丰富的中文注释，希望可以给各位带来阅读上的方便。 基于meanshift的聚类分割方法包括滤波、区域融合等操作，通过调整sigma和sigmar来调整分割效果。(According to D. Comaniciu, P. Meer: Mean Shift: A robust approach toward feature space analysis, and C. Christoudias, B. Georgescu, P. Meer: Synergism in low level vision. These two documents prepared by the methods provided by image segmentation code , the author is Chris M. Christoudias, Bogdan Georgescu, after I read the code, add a rich Chinese notes, hoping to bring you the convenience of reading. Segmentation method based on clustering meanshift including filtering, regional integration and other operations, and by adjusting the sigma sigmar to adjust segmentation results.)

程序将界面与数据处理分开，界面采用MFC的对话框模板，对注册、训练、识别和主界面分别建立一个类，通过消息处理函数与用户进行交互，利用多线程来实时显示图像。 数据处理又分为了CFaceAlign（人脸检测+几何归一化）、CLightPrep（光照归一化）、CFaceFeature（Gabor特征提取）、CSubspace（计算Fisherface子空间）四个类，还有一个类 CFaceMngr 负责管理界面与数据之间的交流。这样的程序结构使对界面和算法彼此独立，互不影响。 每个类的具体方法请见相应头文件中的注释。(The program will interface with the data processing separate interface using MFC dialog template, registration, training, recognition and the main interface, respectively, to create a class that interacts with the user through a message handler, using multiple threads to display images in real time.   Data processing is divided into a CFaceAlign (Face Detection+ geometric normalization), CLightPrep (illumination normalization), CFaceFeature (Gabor feature extraction), CSubspace (computing Fisherface subspace) four categories, as well as a class is responsible for managing CFaceMngr interface and data exchange between. Such a program structure allows for interfaces and algorithms are independent of each other. Specific methods of each class, see the comments in the appropriate header files.)

基于菲涅耳全息图和离散余弦变换的_盲数字水印_技术(Blind digital watermarking _ technology based on the Fresnel hologram and the discrete cosine transform _)