【cv2】：关于opencv-Python的函数

缺乏、安全感 2022-01-12 12:59 577阅读 0赞

## 关于opencv-Python的函数 ##

发布时间： 2019-04-13 19:50:42

Chapter one：图片  
读入图像 cv2.imread(filename\[, flags\]) → retval  
显示图像 cv2.imshow(winname, mat) → None  
cv2.waitKey(\[delay\]) → retval  
创建窗口 cv2.namedWindow(winname\[, flags\]) → None  
摧毁窗口 cv2.destroyWindow(winname) → None  
cv2.destroyAllWindows()→ None  
保存图像 cv2.imwrite(filename,img\[, params\]) → retval

Chapter two: 视频  
读取视频文件、摄像头获取视频 cv2.VideoCapture(filename/device) →  
.read(\[image\]) → retval,image  
.open()→ retval  
.isOpened()→ retval  
视频文件属性、设置 .get(prodId) → retval  
.set(prodId,value) → retval  
保存视频文件 cv2.VideoWriter(\[filename,fourcc,fps,framesize\[,isColor\]\]) →  
.write(image) → None  
关闭视频文件 .release()→ None

Chapter three: 绘图  
[http://wiki.opencv.org.cn/index.php/Cxcore绘图函数\#GetTextSize][http_wiki.opencv.org.cn_index.php_Cxcore_GetTextSize]  
只有C++：InitFont、arrowedLine、

画圆 cv2.circle(img,center,radius,color\[,thickness\[,lineType\[,shift\]\]\]) → None  
画矩阵 cv2.rectangle(img,pt1,pt2,color\[,thickness\[,lineType\[,shift\]\]\]) → None  
画线段 cv2.Line(img,pt1,pt2,color\[,thickness\[,lineType\[,shift\]\]\]) → None  
画折线 cv2.polyLines(img,pts,isClosed,color\[,thickness\[,linrType\[,shift\]\]\]) → None  
剪切矩形内部的直线 cv2.clipLine(imgRect,pt1,pt2) → retval,pt1,pt2  
画椭圆 cv2.ellipse(img,center,axes,angle,startAngle,endAngle,color\[,thickness\[,lineType\[,shift\]\]\]) → None  
cv2.ellipse(img,box,color\[,thickness\[,lineType\]\]) → None  
用折线逼近椭圆弧 cv2.ellipse2Poly(center,axes,angle,arcStart,arcEnd,delta) → pts  
填充多边形 cv2.fillConvexPoly(img,points,color\[,lineType\[,shift\]\]) → None  
填充多边形内部 cv2.fillPoly(img,pts,color\[,lineType\[,shift\[,offset\]\]\]) → None  
在图像中显示文本字符串 cv2.putText(img,text,org,fontFace,fontScale,color\[,thickness\[,lineType\[,bottomLeftOrigin\]\]\]) → None  
获得文本字符串的宽度和高度 cv2.getTextSize(text,fontFace,fontScale,thickness) → retval, baseLine  
画外部和内部的轮廓 cv2.drawContours(image,contours,contourIdx,color\[,thickness\[,lineType\[,hierarchy\[,maxLevel\[,offset\]\]\]\]\]) → None

Chapter four: 处理鼠标事件  
处理鼠标事件 cv2.SetMouseCallback(winname,onMouse,param=None) → None

Chapter five: 滑动条  
创建滑动条 cv2.createTrackbar(trackname,winname,value,count,onChange) → None  
return 滑动条属性 cv2.getTrackbarPos(trackname,winname) → retval  
设置滑动条属性 cv2.setTrackbarPos(trackname,winname,pos) → None

Chapter six:图像操作  
获取图像形状 .shape  
获取像素数目 .size  
获取图像数据类型 .dtype  
拆分、合并通道 cv2.split(m\[,mv\]) → mv  
cv2.merge(mv\[,dst\]) → dst  
图像扩边 cv2.copyMakeBorder(src,top,bottom,left,right,borderType\[,dst\[,value\]\]) → dst  
图像减法 cv2.subtract(src1,src2\[,dst\[,mask\[,dtype\]\]\])→dst  
图像加法 cv2.add(src1,src2\[,dst\[,mask\[,dtype\]\]\]) → dst  
图像混合 cv2.addWeighted(src1,alpha,sec2,beta,gamma\[,dst\[,dtype\]\]) → dst  
图像位运算 cv2.bitwise\_and(src1,src2\[,dst\[,mask\]\]) → dst  
cv2.bitwise\_not(src\[,dst\[,mask\]\]) → dst  
cv2.bitwise\_or(src1,src2\[,dst\[,mask\]\]) → dst  
cv2.bitwise\_xor(src1,src2\[,dst\[,mask\]\]) → dst

Chapter seven: 颜色空间转换  
转换颜色空间 cv2.cvtColor(src,code\[,dst\[,dstcn\]\]) → dst  
检查数组元素是否在两个数量之间 cv2.inRange(src,lowerb,upperb\[,dst\]) → dst

Chapter eight: 几何变换  
[http://www.cnblogs.com/dupuleng/articles/4055020.html][http_www.cnblogs.com_dupuleng_articles_4055020.html]  
扩展缩放 cv2.resize(src,dsize\[,dst\[,fx\[,fy\[,interpolation\]\]\]\]) → dst  
仿射变换 cv2.getAffineTransform(src,dst) → retval  
cv2.warpAffine(src,M,dsize\[,dst\[,flags\[,borderMode\[,borderValue\]\]\]\]) → dst  
旋转 cv2.getRotationMatrix2D(center,angle,scale)→ret  
透视变换 cv2.getPerspectiveTransform(src,dst) → retval  
cv2.warpPerspective(src,M,dsize\[,dst\[,flags\[,borderMod\[,borderValue\]\]\]\]) → dst

Chapter nine:图像阈值  
简单阈值 cv2.threshold(src,thresh,maxvalue,type\[,dst\])→retval,dst  
自适应阈值 cv2.adaptiveThreshold(src,maxvalue,adaptiveMethod,thresholdType,blockSize,\[,dst\]) → dst

Chapter ten: 图像平滑  
2D卷积 cv2.filter2D(src,ddepth,kernel\[,dst\[,anchor\[,delta\[,borderType\]\]\]\]) → dst  
平均 cv2.blur(src,ksize,\[,dst\[,anchor\[,borderType\]\]\])→dst  
cv2.boxFilter(src,ddepth,ksize\[,dst\[,anchor\[,normalize\[,borderType\]\]\]\]) → dst  
高斯模糊 cv2.GuassianBlur(src,ksize,sigmaX\[,dst\[,sigmaY\[,borderType\]\]\]) → dst  
cv2.getGuassianKernel(ksize,sigma\[,kType\]) → retval  
中值模糊 cv2.medianBlur(src,ksize\[,dst\]) → dst  
双边滤波 cv2.bilateralFilter(src,d,sigmaColor,sigmaSpace\[,dst\[,borderType\]\]) → dst

Chapter eleven: 形态学转换  
腐蚀 cv2.erode(src,kernel\[,dst\[,anchor\[,iterations\[,borderType\[,borderType\]\]\]\]\]) → dst  
膨胀 cv2.dilate(src,kernel\[,dst\[,anchor\[,iterations\[,borderType\[,borderType\]\]\]\]\]) → dst  
形态学操作（开运算，闭运算，梯度，礼帽，黑帽） cv2.morphologyEx(src,op,kernel\[,dst\[,anchor\[,iterations\[,borderType\[,borderValue\]\]\]\]\])→ dst  
结构化元素 cv2.getStructuringElement(shape,ksize\[,anchor\])→retval

Chapter twelve: 图像梯度  
Sobel算子 cv2.Sobel(src,ddepth,dx,dy\[,dst\[,ksize\[,scale\[,delta\[,borderType\]\]\]\]\]) →dst  
Scharr算子 cv2.Scharr(src,ddepth,dx,dy\[,dst\[,scale\[,delta\[,borderType\]\]\]\]) → dst  
Laplacian算子 cv2.Laplacian(src,ddepth\[,dst\[,ksize\[,scale\[,delta\[,borderType\]\]\]\]\])→dst

Chapter thirteen: canny边缘检测  
cv2.Canny(image,threshold1,threshold2\[,edges\[,apertureSize\[,L2gradient\]\]\]) → edges

Chapter fourteen: 轮廓  
查找轮廓 cv2.findContours(image,mode,method\[,contours\[,hierarchy\[,offset\]\]\]) → contours,hierarchy  
绘制轮廓 cv2.drawContours(image,contours,contourIdx,color\[,thickness\[,lineType\[,hierarchy\[,maxLevel\[,offset\]\]\]\]\]) → None  
图像的矩 cv2.moments(array\[,binaryImage\]) → retval  
轮廓面积 cv2.contourArea(contour\[,oriented\]) → retval  
轮廓周长 cv2.arcLength(curve,closed) → retval  
轮廓近似 cv2.approxPolyDP(curve,epsilon,closed\[,approxCurvel\])→approxCurvel  
凸包 cv2.convexHull(point\[,hull\[,clockwise\[,returnPoints\]\]\]) → hull  
凸缺陷 cv2.convexityDefects(contour,convexhull\[,convexityDefects\])→convexityDefects  
凸性检测 cv2.isContourConvex(contour) → retval  
边界矩形 cv2.boundingRect(points) → retval  
cv2.minAreaRect(points) → retval  
cv2.boxPoints()  
最小外接圆 cv2.minEnclosingCircle(points) → center,radius  
点到轮廓距离 cv2.pointPolygonTest(contour,pt,measureDist) → retval  
椭圆拟合 cv2.fitEllipse(points) → retval  
直线拟合 cv2.fitLine(points,distType,param,reps,aeps\[,line\]) → line  
形状匹配 cv2.matchShapes(contour1,contour2,method,paramter) → retval  
模块匹配 cv2.matchTemplate(image,temp1,method\[,result\]) → result  
最值及其位置 cv2.minMaxLoc(src\[,mask\]) → minVal,maxVal,minLoc,maxLoc  
平均颜色/灰度 cv2.mean(src\[,mask\]) → retval

Chapter fifteen: 直方图  
绘制直方图 cv2.calcHist(image,channels,mask,histSize,ranges\[,hist\[,accumulate\]\]) → hist  
直方图均衡化 cv2.equalizeHist(src\[,dst\]) → dst  
cv2.createCLAHE(\[,clipLimit\[,tileGridSize\]\]) → retval  
直方图反向投影 cv2.calcBackProject(image,channels,hist,range,scale\[,dst\]) → dst  
归一化 cv2.normalize(src\[,dst\[,alpha\[,beta\[,norm\_type\[.dtype\[,mask\]\]\]\]\]\])→dst

Chapter sixteen: 图像变换  
傅立叶变换 cv2.dft(src\[,dst\[,flags\[,nonzeroRows\]\]\]) → dst  
cv2.getOptimalDFTSize(vecsize) → retvach

Chapter seventeen: 霍夫变换  
霍夫直线变换 cv2.HoughLines(image,rho,theta,threshold\[,lines\[,srn\[,stn\]\]\]) → lines  
cv2.HoughLinesP(image,rho,theta,threshold\[,lines\[,minLineLength\[,maxLineGap\]\]\]) → lines  
霍夫圆环变换 cv2.HoughCircles(image,method,dp,minDist\[,circles\[,param1\[,param2\[,minRadius\[,maxRadius\]\]\]\]\]) → lines

Chapter eighteen: 分水岭算法图像分割  
分水岭算法 cv2.watershed(image,maskers) → None  
连通分支 cv2.connectedComponents(image\[,labels\[,connectivity\[,ltype\]\]\]) → retval,labels

Chapter nineteen: grabcut算法交互式前景提取  
cv2.grabcut(image,mask,rect,bgdModel,fgdModel,iterCount\[,mode\]) → None

Chapter twenty: 角点检测  
C++:DescriptorExtractor\_create()  
Harris角点检测 cv2.cornerHarris(src,blockSize,ksize,k\[,dst\[,borderType\]\]) → dst  
提炼角位置 cv2.cornerSubPix(image,corners,winSize,zeroZone,criteria) → None  
Shi-Tomasi角点检测 cv2.goodFeaturesToTrack(image,maxCorners,qualityLevel,minDistance\[,corners\[,mask\[,blockSize\[,useHarrisDetect\[,k\]\]\]\]\]) → corners  
SURF cv2.SURF(\[hessianThreshold\[,n0ctaves\[,n0ctaveLayers\[,extend\[,upright\]\]\]\]\]) →  
SURF检测和计算要点 .detect(image\[,mask\]) → keypoints  
.detectAndCompute(image,mask\[,descriptors\[,useProvidedKeypoints\]\]) → keypoints,descriptors  
FAST特征检测 cv2.FastFeatureDetector\_create(\[,threshold\[,nonmaxSuppression\[,type\]\]\]) → retval  
绘制要点 cv2.drawKeypoints(image,keypoints,outImage\[,color\[,flags\]\])→outImg

Chapter twenty-one: 特征匹配  
C++: FLANN  
Brute-Force cv2.BFMatcher(\[,normType\[,crossCheck\]\]) →  
对象透视图变换 cv2.findHomography(srcPoints,dstPoints\[,method\[,ransacReprojThreshold\[,mask\]\]\]) → retval,mask

Chapter twenty-two: 在视频中找到并跟踪目标对象  
Meanshift cv2.meanShift(proImage,window,criteria) → retval,window  
camshift cv2.camShift(proImage,window,criteria) → retval,window

Chapter twenty-three: 光流  
Lucas-Kanade光流 cv2.calcOpticalFlowPyrLK(prevImg,nextImg,prevPts\[,nextPts\[,status\[,err\[,winSize\[,maxLevel\[,criteria\[,derivLambda\[,flags\]\]\]\]\]\]\]\])  
→nextPts,status,err  
cv2.goodFeaturesToTrack(image,maxCorners,qualityLevel,minCorners\[,corners\[,mask\[,blockSize\[,useHarrisDetector\[,k\]\]\]\]\]) → corners  
GF稠密光流 cv2.calcOpticalFlowFarneback(prevImg,nextImg,pyr\_scale,levels,winsize,iterations,poly\_n,poly\_sigma,flags\[,flow\]) → flow

Chapter twenty-four: 背景减除  
BackgroundSubtractorMOG  
BackgroundSubtractorGMG

BackgroundSubtractorMOG cv2.BackgroundSubtractorMOG(history,nmixtures,backgroundRatio\[,noiseSigma\]) →  
createBackgroundSubtractorMOG2(\[, history\[, varThreshold\[, detectShadows\]\]\]) → retval

Chapter twenty-five: 摄像机设定和3D重构  
将对象点转换到图像点 cv2.projectPoints(objectPoints,rvec,tvec,cameraMatrix,distCoeffs\[,imagePoints\[,jacobian\[,aspectRatio\]\]\])  
→ imagePoints,jacobian  
摄像机标定 cv2.calibrateCamera(objectPoints,imagePoints,imageSize\[,cameraMatrix\[,distCoeffs\[,rvecs\[,tvecs\[,flags\]\]\]\]\])  
→ retval,cameraMatrix,distCoeffs,rvecs,tvecs  
畸变校正 cv2.getOptimalNewCameraMatrix(cameraMatrix,distCoeffs,imageSize,alpha\[,newImgSize\[,centerPrincipalPoint\]\])  
→retval,validPixROI  
cv2.undistort(src,cameraMatrix,distCoeffs\[,dst\[,newCameraMatrix\]\]) → dst  
cv2.remap(src,map1,map2,interpolation\[,dst\[,borderMode\[,borderValue\]\]\]) → dst  
计算旋转和变换 cv2.solvePnPRansac(objectPoints,imagePoints,cameraMatrix,distCoeffs\[,rvec\[,tvec\[,useExtrinsicGuess\[,iterationsCount\[,reprojectionError\[,minInliersCount\[,inliers\[,flags\]\]\]\]\]\]\]\])  
→ rvec,tvec,inliers

Chapter twenty-six: 机器学习  
K近邻 cv2.KNearest.train(trainData,responses\[,sampleIdx\[,isRegression\[,maxK\[,updateBase\]\]\]\]) → retval  
cv2.KNearest.find\_nearest(samples,k\[,results\[,neighborResponses\[,dists\]\]\]) → retval,results,neighborResponses,dist  
SVM机 cv2.SVM(trainData,responses\[,varIdx\[,sampleIdx\[,params\]\]\]) →  
.train(traindata,responses\[,varIdx\[,sampleIdx\[,params\]\]\]) → retval  
.train\_auto(trainData,responses,varIdx,sampleIdx,params\[,k\_fold\[,Cgrid\[,gammaGrid\[,pGrid\[,nuGrid\[,coeffGrid\[,degreeGrid\[,balanced\]\]\]\]\]\]\]\]) → retval  
.predict(sample\[, returnDFVal\]) → retval  
.save()  
K值聚类 cv2.kmeans(data,K,criteria,attempts,flags\[,bestLabels\[,centers\]\]) → retval,bestLabels,centers

Chapter twenty-seven: 计算摄影学  
图像去噪 cv2.fastNlMeansDenoising(src\[,dst\[,h\[,templateWindowSize\[,searchWindowSize\]\]\]\])→dst  
cv2.fastNlMeansDenoising(src,h\[,dst\[,templateWindowSize\[,searchWindowSize\[,normType\]\]\]\])→dst  
fastNlMeansDenoisingMulti(srcImgs,imgToDenoiseIndex,temporalWindowSize\[,dst\[,h\[,templateWindowSize\[,searchWindowSize\]\]\]\])→dst  
cv2.fastNlMeansDenoisingMulti(srcImgs,imgToDenoiseIndex,temporalWindowSize,h\[,dst\[,templateWindowSize\[,searchWindowSize\[,normType\]\]\]\])→dst  
cv2.fastNlMeansDenoisingColored(src\[,dst\[,h\[,hColor\[,templateWindowSize\[,searchWindowSize\]\]\]\]\])→dst  
fastNlMeansDenoisingColoredMulti(srcImgs,imgToDenoiseIndex,temporalWindowSize\[,dst\[,h\[,hColor\[,templateWindowSize\[,searchWindowSize\]\]\]\]\])→dst  
图像修补 cv2.inpaint(src,inpaintMask,inpaintRange,flags\[,dst\]) →dst

Chapter twenty-eight: 使用Haar 分类器进行面部检测  
[https://www.wandouip.com/t5i199888/][https_www.wandouip.com_t5i199888]

[http_wiki.opencv.org.cn_index.php_Cxcore_GetTextSize]: http://wiki.opencv.org.cn/index.php/Cxcore%E7%BB%98%E5%9B%BE%E5%87%BD%E6%95%B0#GetTextSize
[http_www.cnblogs.com_dupuleng_articles_4055020.html]: http://www.cnblogs.com/dupuleng/articles/4055020.html
[https_www.wandouip.com_t5i199888]: https://www.wandouip.com/t5i199888/