quyushengzhangfa

改程序是用来行人检测的，有源代码，可以轻松学习并参考。(The reform program is used for pedestrian detection, source code, you can easily learn and reference.)

通过非监督分类中的K-均值聚类对Indian pines高光谱图像进行分类(Classification of hyperspectral images of pines Indian by K-means of non supervised classification)

结构相似度函数，数字图像质量评价的一种经典方法(for image quality evaluation)

对遥感影像或者图形图像，是在摄影测量数据处理中比较知名的Moravec算子。希望可以帮助到大家！(Remote sensing image or graphic images, photogrammetric data processing in the more well-known Moravec operator. I hope you can help to everyone!)

通过区域迭代合并与定位切图的图像分割算法，非常好的。(Image Segmentation by Iterated Region Merging with Localized Graph Cuts)

说明：  在VC中实现了使用Moravec算子对遥感影像进行特征点的提取，缺点是速度较慢。不过提取效果不错！(Implemented in VC using Moravec operator of remote sensing images feature point extraction, the disadvantage is slow. However, extracting good results!)

运用harr特征+SRC（稀疏表示）分类实现的一种车辆检测方法，文件中提供了训练和测试车辆图片。由于时间原因，所用haar特征没有优化，维度过高，导致滑窗框图过慢，本代码只输出效果统计数据，以供大家参考学习稀疏表示在车辆检测中的应用。(Using harr feature+SRC (sparse representation) classification to achieve a vehicle detection method, the paper provides a training and test vehicle picture. Due to time reasons, the use of haar feature is not optimized, high dimension, resulting in sliding sash figure is too slow, the effect of the code only output statistics for your reference learning sparse representation in the vehicle detection.)

说明：  生成手写数字图片，一共提供了三种算法，example_1和example_2训练GAN来生成手写数字图片，搭建了一个简单的GAN的网络结构，example_3难度较大(There are three algorithms to generate handwritten digital pictures. Example_1 and example_2 train GAN to generate handwritten digital pictures, and build a simple network structure of GAN. Example_3 is more difficult.)

demons算法用来实现图像的配准，特别是医学图像的配准，已经达到了很高的精度(demons algorithm is used to achieve image registration, especially in medical image registration, has reached a very high accuracy)

利用灰度积分投影直接对人脸图像进行检测和眼睛定位是一种常用的算法,但是直接采用该算法会受到背 景、特征等因素的影响,识别准确率较低。提出了一种基于最大类间方差阈值和区域膨胀相结合的检测与定位算法。该算 法首先计算最大类间方差设置阈值,把灰度图像转换为二值图像并检测出人脸区域,然后通过对该人脸区域中的连通区域 进行膨胀及连通性处理,精确定位眼睛坐标。实验表明,此算法可靠,具有较好的识别效果。(face recognized)