GaborTextureSegment

说明：  用matlab写的t-snake的算法用于处理比较简单的情况有一定参考价值(using Matlab write t-snake algorithm for handling simple reference of a certain value)

自己整理的经典边缘检测算子matlab源程序 图像处理必备 1、canny算子 2、kirsch算子 3、laplacian算子 4、log算子 5、prewitt算子 6、robert算子 7、robinson算子 8、sobel算子(Own finishing classical edge detection operator matlab source image processing necessary 1, canny operator, kirsch operator, Laplacian operator, and the log operator Operators 6 prewitt, robert. Operator, robinson operator 8, sobel operator)

高斯混合模型的源代码。从文件中读取数据，用三个高斯混合模型进行处理(Gaussian mixture model of the source code. Read from the file data, using three Gaussian mixture model for processing)

MATLAB做的运动估计，用于图像处理中运动轨迹追踪（仅供大家参考)(motion estimation by matlab, for motion tracking in digital image processing)

图像直方图统计，并生成灰度图。 matlab程序(Histogram of the image statistics, and generate grayscale. Matlab program)

说明：  PCA图像降噪新算法-LPG PCA，全名叫做Two-stage Image Denoising by Principal Component Analysis with Local Pixel Grouping。内含算法代码和数据集，均经本人实测，可进行灰度图像和RGB图像的降噪，且PSNR和SSIM指标明显好于小波滤波、K-SVD等经典算法。值得学习图像降噪的看看！(The new algorithm of PCA image denoising LPG PCA, full name is two stage image denoising by principal component analysis with local pixel grouping. The algorithm code and data set can be used to reduce the noise of gray image and RGB image, and the PSNR and SSIM are better than the classical algorithms such as wavelet filtering and K-SVD. It's worth learning about image denoising!)

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.)

用lbp实现纹理特征提取，并分类说明: 一共有三个m文件，一个是lbp.m, 存放主要的lbp算法， 一个是getmapping,用以做算法的辅助函数， 一个是lbptest.m,存放着测试代码。 这三个文件需要放到同一个文件夹，并在文件夹中添加相应的图片， 具体的图片名字见lbptest.m的代码，运行lbptest.m可以查看结果。代码最后给出效果图 这三个文件是最传统的LBP方法，有256种。(With lbp achieve texture feature extraction and classification Description: A total of three m files, one is lbp.m, store the main lbp algorithm, One is getmapping, helper functions for doing arithmetic, One is lbptest.m, kept the test code. These three documents need to put the same folder, and add the appropriate folder, pictures, See lbptest.m specific name of the picture of the code, run lbptest.m can view the results. Finally renderings code These three documents are the most traditional LBP method, there are 256 kinds.)

各种边缘检测程序，包括2Sobel算子边缘检测， Prewitt算子边缘检测，Roberts算子边缘检测， Laplace算子边缘检测，Canny算子边缘检测 (Edge Color)

说明：  基于标记的分水岭算法，弥补了传统分水岭算法存在的过分割问题。(Marker-based watershed algorithm makes up for the over-segmentation problem of traditional watershed algorithm.)