computer-vision
matdodo
use knowledge to make the world better.
展开
-
[cv] linearity and convolution
Linearity:An operator or system H is linear if two properties hold:1. Additivity: H(f1+f2) = H(f1) + H(f2)2. scaling multiplicative H(a*f1) = a*H(f1)where f1, f2 are functions, a is原创 2017-05-05 23:31:33 · 550 阅读 · 0 评论 -
[cv] optical flow
Given optical flow, we can detect moving objects, detect how far the objects are, and detect independent moving objects.To overcome this problem, there are two solutions provided.原创 2017-08-15 19:23:21 · 542 阅读 · 0 评论 -
[cv]Image transformation
Image transform原创 2017-08-13 01:10:52 · 674 阅读 · 0 评论 -
[cv] Feature detection
Gabor Function2d - gaborSIFTHOG原创 2017-08-04 14:46:24 · 443 阅读 · 0 评论 -
[cv] Image Convolution and Edge Detection (2nd)
Non-max supression:Edge Linking原创 2017-08-02 17:16:26 · 721 阅读 · 0 评论 -
[cv] Fourier Transform, Image Convolution, Edge Detection
1. Fourier Transformconvolution neutral networks is a hot spot in machine learning technique.But is that convolution the same as the following convolution?let us start with 1D signal.原创 2017-07-13 00:03:44 · 963 阅读 · 0 评论 -
[cv] camera geometry and color sensing
camera geometry and color sensing原创 2017-07-10 22:18:55 · 537 阅读 · 0 评论 -
visual studio 2017 配置 opencv 3.2 开发环境
1. 安装opencv 3.2链接地址:https://sourceforge.net/projects/opencvlibrary/files/opencv-win/3.2.0/opencv-3.2.0-vc14.exe/download下载后双击进行解压。2. 配置visual studio 2017文件->新建->项目在源文件处右键选择, 添加, 新建项,原创 2017-06-07 12:33:47 · 7340 阅读 · 1 评论 -
[cv]generalized hough transform
general houghspecific hough transform is usually used in images with a specific shape (line, circle). examplealgorithm原创 2017-05-10 01:44:25 · 1959 阅读 · 0 评论 -
[cv]hough transform ----circle
hough transform for circlesif the radius is not fixed, hough of circle in image will be a 3D space. but if the gradient of a circle is given, the search space is decreased heavily. tips pros and c原创 2017-05-10 01:15:16 · 915 阅读 · 0 评论 -
[cv] convolution in frequency domain
convolution in frequency domainconvolution in spatial domain == multiplication in frequency domain.FFT Fast Fourier Transformsmoothing and blurlow and high pass filter properties of Fourier TransformFo原创 2017-05-17 21:12:43 · 923 阅读 · 0 评论 -
[cv]edge detection: 2d operators
derivative of 2d gaussian filterGaussian derivative in x,y directioneffect of sigma in smoothing filtercanny edge operator in matlab, we can use command: edge(image, ‘canny’) easily. In canny opera原创 2017-05-09 00:30:15 · 1423 阅读 · 0 评论 -
[cv]edge detection: gradients
usually, filter is used to find a specific pattern in the pictures. If we want to find lines and edges in the picture, firstly we think about gradients. 画图的时候,光画出线条就可以表达很多信息。 不要轻视线条,其中包含了太多的信息。edge原创 2017-05-08 18:23:57 · 812 阅读 · 0 评论 -
[cv] filters as templates-normxcorr2(template,img)
introductionimages are not just functions (intensity). it also contains location properties.1D correlation the filter is normalized and then is used to compute cross -correlation. why the peak value原创 2017-05-07 22:30:32 · 1608 阅读 · 0 评论 -
[cv][dsp] filter
filter classificationFilter stablility原创 2017-09-23 04:20:16 · 549 阅读 · 0 评论