shuixia

基于Powell和互信息的多分辨率图像配准(Image registration Powell and mutual information)

本代码建立高斯混合模型(高斯多模型)(GMM)，将其用于计算机视觉领域的视频目标检测/视频监控/运动检测/运动目标检测/视频目标跟踪等相关应用中。(This code sets up Gaussian mixture model(Gaussion Mixture Model, GMM) with image sequences for the research and application field of computer vision, using GMM for visual object detection(video object detection), visual surveillance, motion detection, motion tracking, visual object tracking, video object tracking, etc.)

python实现的三维空间维罗尼图算法演示。(3d voronoi use python)

说明：  利用K均值聚类对图像lena进行分割，图片可根据需求进行更换。(Image Lena is segmented by k-means clustering, and the image can be replaced according to the demand.)

数学建模，动态模糊与模糊图像处理，维纳滤波方法。(Mathematical modeling, dynamic fuzzy and fuzzy image processing, Wiener filtering method.)

说明：  QR二维码识别中在对图像预处理后最重要的一步就是要进行定位，QR二维码中有三个位置探测图形，通过扫描其特征便可以进行精细定位，提出了一种新的算法用于筛选候选点。(In QR code recognition, the most important step after image preprocessing is to locate. There are three position detection patterns in QR code, which can be precisely located by scanning its features. A new algorithm is proposed to screen candidate points.)

直方图拉伸是通过对比度拉伸对直方图进行调整，从而“扩大”前景和背景灰度的差别，以达到增强对比度的目的(Histogram stretching histogram is adjusted by contrast stretching, thereby " expand" the difference between the foreground and background of gray, in order to achieve the purpose of enhancing the contrast)

该程序为离焦模糊参数识别的方法,通过canny算子检测边缘，用hough变换检测直线的方法求离焦半径。(The program for the defocus blur parameter identification method by canny operator edge detection using hough transform line detection method for the defocus radius.)

在图像中加入某个分数值（int），然后检测这个分数，要求在图像受到攻击时能有效（100％）提取这个值(In the image by adding a Score value (int), then detection of the scores, the requirements in the image can be attacked effectively (100) extraction of this value)

区域生长算法，通过设置种子点来实现,代码注释完整(region grow algorithm with matlab)