石鑫华视觉网-中科院自动化所 计算机视觉课件

polsarpro官方学习教程，系统的说明了polsarpro软件的操作使用，可进行树高反演，地物分类，已经森林模拟数据生成(polsarpro tutorial note)

这是香港大学沙威博士写的压缩感知引论及相应的OMP算法。(This is written by the University of Hong Kong, Dr Savit compressed sensing cited dealt with the OMP algorithm.)

课程设计 基于matlab的工件图像尺寸测量技术(The course design is based on matlab the workpiece image size measurement techniques)

说明：  SIFT，即尺度不变特征变换（Scale-invariant feature transform，SIFT），是用于图像处理领域的一种描述。这种描述具有尺度不变性，可在图像中检测出关键点，是一种局部特征描述子。 该方法于 1999 年由 David Lowe 首先发表于计算机视觉国际会议（International Conference on Computer Vision，ICCV），2004 年再次经 David Lowe 整理完善后发表于 International journal of computer vision（IJCV） 。截止 2014 年 8 月，该论文单篇被引次数达 25000 余次。(来自百度)(SIFT, namely Scale-invariant feature transform (SIFT), is a description used in the field of image processing. This description has scale invariance and can detect key points in images. It is a local feature descriptor. This method was first published by David Lowe at the International Conference on Computer Vision (ICCV) in 1999 and published in the International Journal of Computer Vision (IJCV) in 2004 after it was sorted out and perfected by David Lowe again. As of August 2014, more than 25,000 citations have been cited. (From Baidu))

压缩感知 图形图像 图像处理 数据恢复 代码(Data recovery of compressed sensing image image processing)

高光谱遥感图像光谱解混的独立成分分析技术(Hyperspectral remote sensing image spectra unmixing independent component analysis)

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)

这段代码的功能是实现图像去雾处理，Matlab代码，可以直接运行的(Functionality of the code is to realize image processing demisters Matlab code can be run directly)

根据国外文献算法中所写的MATLAB代码，实现肤色检测与轮廓提取。(According to foreign literature written in MATLAB code algorithm to achieve skin color detection and contour extraction.)

此代码为传统的snake模型算法代码，运行其中的egsnake.m即可。该代码实现对图像的边缘检测，仅供大家一块学习交流(This code is the code algorithm of traditional snake model, which can run egsnake.m. The code to achieve the image edge detection, we only a learning exchange.)