zhongxin

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.)

说明：  对于双基地SAR，使用BP算法进行成像，展示场景、回波信号、距离压缩后以及最后成像的图像。(For Bistatic SAR, BP algorithm is used to image the scene, echo signal, range compression and the final image.)

《基于计算机视觉的普适性三维重建方法研究》源代码，实现了一种将二维图像重建成三维场景的方法。(Source code of <General Sense Three-Dimension Reconstruction Research Based on Computer Vision>, a technique which can convert 2D images into 3D sence.)

图像分割matlab代码，区域生长，效果很好(Image segmentation matlab code, region growing)

chan和vese的多水平集算法，可以用2条初始水平集来演化，从而达到检测多目标的目的。(chan and the multi-level set algorithm, you can use two sets to the initial level of evolution, so as to achieve the goal of multi-purpose.)

感兴趣区域提取，能用于车牌号及台风等图像，效果不错(Extract the interesting area, it can be used in license plate number and typhoons images and so on, has good results )

有关图像的目标识别："给出一种基于特征分类辨识的合成孔径雷达图像目标检测方法#用恒虚警和扩展分形方法对3&E图像进行目 标检测后用面积和峰值能量比算子辨识目标和背景杂波!去除一部分虚警!用小波域主成分分析对每个检测窗口内的图 像提取特征向量!用支持向量机对提取得到的特征向量进行分类!辨识目标和背景杂波!完成目标检测#使用&K?3数 据对该方法进行验证和分析!实验结果表明!经过特征分类辨识后!在检测率不变的情况下!虚警数目显著降低# (Related to the image target recognition)

手指静脉预处理，图像分割，静脉纹路分割，滤波去噪，细化等相关代码(Finger vein pretreatment, image segmentation, vein segmentation, filtering and denoising, thinning and other related codes)

Car Plate recognition for trail for learn

基于扫描地形图的几何纠正算法研究。主要讲解了关于倾斜的数字图像的处理问题。(Scanned topographic maps based on the geometric correction algorithm. Mainly on the inclination to explain the deal with the issue of digital images.)