matlab

基于mind相似性度量的图像配准，配准精度很高。效果不错。(Mind based on similarity measure of image registration, the registration precision is high. The result is right.)

一个人体运动识别跟踪的示例程序，里面附带了一个运动视频，运行run.m文件。程序里包含了轮廓提取，跟踪算法，很好的tracking学习程序(A human motion recognition tracking sample program, which comes with a sports video, to run run.m file. The program includes a contour extraction and tracking algorithm, good tracking learning program)

图象处理用于对BMP格式的图象进行几何校正.(image processing used to BMP format images geometric correction.)

sift算法在cv领域的重要性不言而喻，该作者的文章引用率在cv界是number1.本篇博客只是本人把sift算法知识点整理了下，以免忘记。本文比较早的一篇博文opencv源码解析之(3)：特征点检查前言1 中有使用opencv自带的sift做了个简单的实验，而这次主要是利用Rob Hess的sift源码来做实验，其实现在的opencv版本中带的sift算法也是Rob Hess的，只是稍微包装了下。 下面来做下试验,试验sift代码采用Rob Hess的代码，opencv目前版本中的sift源码也是采用Rob Hess的。代码可以在他的主页上下载：http://blogs.oregonstate.edu/hess/code/sift/ 这里我下载的是windows版本的，并采用Qt做了个简单的界面。(sift algorithm is self-evident the importance of the cv field, the author of the article referenced in the cv sector is number1 this blog I just sift algorithm knowledge compiled under, so as not to forget. This article compares the earlier blog post opencv source code analysis (3): check the preface of the feature points 1 the opencv bring their own sift to do a simple experiment, but this time Rob Hess' s sift source to experiment. In fact, with the opencv version sift algorithm is Rob Hess, only slightly packing the next. Below to do the next test, test sift code using the code of Rob Hess, sift source opencv version of Rob Hess. The code can be downloaded on his home page: http://blogs.oregonstate.edu/hess/code/sift/ here I downloaded the windows version, and using Qt to do a simple interface.)

说明：  为了在抑制噪声时尽可能保留图像边缘信息，提出一种自适应各向异性高斯滤波器设计方法. 该方法由独立 强度传播（IDS）模型决定滤波器的长轴尺度，由像素的邻域平滑度决定长短轴的比例，然后根据该处的灰度梯度方 向自适应决定各向异性滤波器的长轴方向. 仿真实验表明，提出的自适应各向异性滤波器具有很强的噪声抑制和 边缘保持能力. (In order to suppress noise, as far as possible to keep the image edge information, an adaptive anisotropic Gaussian filter design methods. The method by the independent strength spread (IDS) model determines the long axis of the filter scale, the pixel neighborhood smoothness determine the ratio of minor axis, and then there' s gray gradient direction of adaptive decision the long axis of anisotropic filter. Simulation results show that the proposed adaptive anisotropic filter has a strong noise suppression and preserve edges.)

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

7670 32F4程序 代码 适合STM32F4探索者(7670 32F4 program code suitable for STM32F4 Explorer)

对特征值进行非极大值抑制的角点检测，并且选用了合适的阈值加强检测效果(corner detection by non-maxima suppression of Eigenvalues and selected the appropriate threshold to enhance detection)

图像匹配的实现，其中match.m：测试程序，sift.m :尺度不变特征变换（SIFT算法）的核心算法程序，appendimages.m: 该函数创建一个新的图像分别包含两个匹配的图像和他们之间的匹配对的连接直线，可以运行。 (The realization of image matching, which match.m: test program, sift.m: scale-invariant feature transform (SIFT algorithm) of the core algorithm, appendimages.m: This function creates a new image each consists of two images and matching match between them on the connection line, you can run.)

画roc曲线的matlab程序，roc曲线又称操作特性曲线 程序注释很详细(Roc curve matlab painting process, roc curve, also known as operating characteristic curve procedures in great detail the Notes)