sfla

说明：  两种基于非下采样变换图像去噪方法，效果比contourlet变换要好(Two transform based image denoising nonsubsampled method, the effect is better than contourlet transform)

说明：  最小噪声分离变换（Minimum Noise Fraction Rotation，MNF Rotation）工具用于判定图像数据内在的维数（即波段数），分离数据中的噪声，减少随后处理中的计算需求量(Minimum Noise Fraction Rotation (MNF Rotation) tool is used to determine the intrinsic dimension (i.e. band number) of image data, separate the noise in data, and reduce the computational requirements in subsequent processing.)

说明：  随机森林的行人检测。针对视频和图像中快速、准确的行人检测问题,提出了一种分层次的、全局信息和局部信息相结合的行人检测算法。该方法以随机森林分类器为基础。(Pedestrian detection based on random forest)

基于纹理分析的图像分割用滤波实现，需要先输入参数。％The Gabor filter is basically a Gaussian (with variances sx and sy along x and y-axes (Based on Texture Analysis for Image Segmentation with filtering to achieve, need to input parameters. The Gabor filter is basically a Gaussian (with variances sx and sy along x and y-axes)

说明：  基于区域的分裂合并图像分割方法，包含三个m文件，predicate是分裂合并准则代码，split_test用于检测是否需要分裂，splitmerge用于分裂合并(Region Based Split Merge image segmentation method, including three m files, predict is the splitting and merging rule code, split_ Test is used to detect whether splitting is needed, and splitmerge is used for splitting and merging)

实现纹理模式的LBP特征表示及分类。 实现一种基于局部二值模式LBP（Local Binary Pattern）的多分辨率灰度尺度及旋转不变性的纹理分类方法(LBP texture model to achieve that as well as the breakdown characteristics. The realization of a model based on local binary LBP (Local Binary Pattern) Multiresolution gray-scale and rotation invariant texture classification)

图像纹理提取gabor小波提取特征一幅输入的图形进行特征提取基于gabor小波变换(image texture extraction of edge detection)

快速TV算法用于图像去模糊、图像去噪等，其中TV模式包括两种(TV went fast algorithm for image blur, image denoising, which includes two TV mode)

极坐标图像转直角坐标图像，相当于球面的展开效果，最好是正方形图像，但是长方形也可以。(POLARTOIM converts polar image to rectangular image)

用于图像分割，基于fcm算法的图像分割，实现模糊均值聚类的分割，更准却的实现分割。(For image segmentation, image segmentation based on FCM algorithm, the realization of fuzzy clustering segmentation, more accurate but to achieve segmentation.)