image-segmation-shengzhang2

实现了对误拍角度的矫正，基于轮廓提取以及旋转矩阵变换实现图像错误识别和矫正(It realizes the correction of misalignment angle, and realizes image error recognition and correction based on contour extraction and rotation matrix transformation.)

人头检测 通过颜色分割产生人头头发部分的二值图像，计算此部分的面积和长宽比来判断是否有人头(human detection)

将一幅M×N的灰度图像用3×3平均滤波器进行模糊，分别再加上一定的高斯 噪声和均匀噪声。然后，设计一个维纳滤波器对这两幅图像进行复原，分别 计算这两幅图像复原前后的PSNR。(Gray-scale image of a MN 33 average filter is vague, respectively, plus a Gaussian noise and uniform noise. Then, design a Wiener filter to recover these two images, the two images before and after the recovery of PSNR were calculated.)

实现对一幅图像的汉字分割，利用图像点扫描(To achieve an image of the Chinese character segmentation, the use of the image spot scanning)

多类图像特征结合上下文信息进行目标识别（分类、分割）(Many types of image features combined with context information for target identification (classification, segmentation))

这是一个实现了发表在CVPR2006的抠图算法。(This package contains an implementation of the matting algorithm described in the CVPR2006 paper.)

虹膜识别图象归一化处理，把环形虹膜区域归一为一个统一的矩形图象，方便下一步特征提取和匹配。(Iris recognition normalized image processing, the annular iris region normalized to a uniform rectangular images, the next step to facilitate feature extraction and matching.)

说明：  在图像分割中，利用Hough变换进行直线提取的matlab源程序。提取方法不同于一般的边缘检测算法。(In image segmentation using Hough transform for straight line extraction matlab source. Extraction method is different from the general edge detection algorithm.)

VC++图像平滑处理+程序源码，实现图象的平滑（去噪声），锐化功能，测试时请在C盘下放一个测试用的Bmp图片，路径：C: est.bmp，没有的话没法测试。如上图所示是水平平滑处理，可以对比下，不过感觉颜色失真了，有兴趣的看一下。命令行编译过程如下： 　　vcvars32 　　rc bmp.rc 　　cl smooth.c bmp.res user32.lib gdi32.lib(VC++ image smoothing+ program source code, to achieve a smooth image (to noise), sharpening function in the C drive when testing a test of decentralization Bmp picture, path: C: test.bmp, no words can not test . Is shown above the level of smoothing, you can compare the next, but the feeling of color distortion, interested look. Command-line compiler as follows: vcvars32 rc bmp.rc cl smooth.c bmp.res user32.lib gdi32.lib)

主分量分析对SAR图像目标进行特征提取，用最近临方法进行分类(Principal component analysis（PCA） of SAR image target feature extraction, classification using nearest-neighbor method)