搜索资源列表
CReadHoughCircle
- 可以提取图像上的圆形区域,比较好上手的opencv程序(You can extract the circular area of the image and compare the opencv program with good hands)
CReadJiaodian
- 可以实现凸显中的角点提取,有利于进一步提取图像特征方便锦衣部的识别,可以比较好的上手。(It can be used to extract the corner points, which can further extract the image features and facilitate the identification of the clothing department.)
dollar
- 基于$1和opencv机器学习方法的图像识别(Image recognition based on machine learning method based on $1 and opencv)
Image grout
- 使用c#的openCV库实现图片的灰度化和二值化,简单明了(Using c # openCV library to realize the grayscale and binary of the image, simple and clear)
MQL_OpenCV4
- 通过提取图像里特征,进行聚类分析,从而可以判断天气情况。(Through the extraction of features in the image, the cluster analysis can be used to judge the weather conditions.)
OpencvPython
- Program for real time tracking of faces using dlib
Source code
- 在opencv上实现双目测距主要步骤是: 1.双目校正和标定,获得摄像头的参数矩阵: 进行标定得出俩摄像头的参数矩阵 cvStereoRectify 执行双目校正 initUndistortRectifyMap 分别生成两个图像校正所需的像素映射矩阵 cvremap 分别对两个图像进行校正 2.立体匹配,获得视差图: stereoBM生成视差图 预处理: 图像归一化,减少亮度差别,增强纹理 匹配过程: 滑动sad窗口,沿着水平线进行匹配搜索,由于校正后左右图片平行,左图
main.cpp
- 图像操作中比较基础的就是透视变换,初进open cv,实现了透视变换,代码如压缩吧所示(Perspective transformation by opencv)
新建 Microsoft Word 文档
- opencv实现傅里叶变换,傅立叶变换是把图像从空间域转化到频率域的变换。空间域:一般的情况下,空间域的图像是f(x,y)=灰度级(0-255),形象一点就是一个二维矩阵,每个坐标对应一个颜色值。频率域:频率:对于图像来说可以指图像颜色值的梯度,即灰度级的变化速度,幅度:可以简单的理解为是频率的权,即该频率所占的比例:能量=幅度(可能不太准确),变换结果为F(u,v),F代表幅度值,u代表x方向的频率,v代表y方向的频率(Fu Liye transform is the transformati
DispScroll
- 采用opencv浏览图像,带滚动条,可显示大图像。(Using OpenCV to browse images, with scrollbars, can display large images.)
颜色直方图C++ opencv代码
- 把图像转换为直方图显示到桌面上(Convert the image to a histogram to display on the desktop)
1RawToBmpgr
- 基于openCV。 读取raw图像,并将raw格式到bpm格式转换的源代码。(Based on openCV. Read the raw image and convert the raw format into the source code of the BPM format.)
WeChat_jump
- 电脑模拟按键实现微信跳一跳自动跳跃。首先就是在上传的autojump.png图片上进行模板匹配,匹配出小人,并计算小人的坐标;如上图所示,框出来的用红色箭头所指即为小人的坐标! 然后就是通过Canny()函数进行图像的边缘检测,这里使用的阈值为5,10基本可以检测出所有边缘信息(Computer simulation keys to achieve WeChat jump one jump automatically. First of all, template matching is per
harrisdemo
- 对图像进行harris特征点提取,代码是用opencv和C++书写(Image harris feature extraction, the code is written with opencv and C ++)
HSV_CIRCLE
- opencv 读取图像像素值到数组,用指针方式实现(opencv read the pixel of pictures and saved.)
main
- 在VS2015环境下基于opencv(计算机视觉开源库)开发的针对图像中感兴趣区域的轮廓提取(Contour extraction for regions of interest in images based on OpenCV (open source library of computer vision) under the VS2015 environment)
PointCloudScanner
- 在vs中正确配置好pcl和opencv,能够将通过kinect2获得的rgb图像和深度信息合成一张点云ply或者pcd文件,可以通过pcl查看。(Correctly configuring PCL and opencv in VS, we can synthesize a point cloud ply or PCD file from RGB image and depth information obtained by kinect2, which can be viewed through
carOpencv3.2
- 这个只是单纯的 通过对阀值还有 形态学滤波 +轮廓查找 筛选 的方式 对车牌进行 识别,特别是形态学滤波 对车牌的识别结果影响很大,这个只是找车牌的一种 普通方法,但对于初学 图像处理的图像来说 可以通过该项目去熟悉 图像处理的一般步骤 还有熟悉一些简单的图像处理方法。(This is simply based on the threshold and morphological filtering and contour search screening way on the license
simulate
- 形状是由图像的轮廓形成的,所以理论上形状识别是通常在边缘或轮廓检测后的步骤。(edge_based_matching The shape is formed by the outline of the image, so theoretically the shape recognition is usually the step after the edge or contour detection)
9:解决像素点生成后,需要手动操作的问题
- 对一副图像进行预处理,然后提取轮廓,再将轮廓进行像素点抽取。(The image is preprocessed, then the contour is extracted and the contour is extracted by pixel.)