Share
Visit

Thursday, February 25, 2016

Opencv 3.1 Tutorial Optical flow (calcOpticalFlowFarneback)

Farneback Optical flow

Opencv simple C++ tutorial and code to achieve optical flow and farneback optical flow of moving object in opencv video. Lets checkt the  video example and the achieved result on my blog. Working and well describe code is included. 

optical flow Farneback

Optical Flow Farneback parameters remarks

You need 2 images at least to calculate optical flow, the previous image (prevgray) and current image (img). 
!! The previous image must be initialized first !! 
Both images have to be gray scale. 

Result is computer in flowUmat which has same size as inputs but format is CV_32FC2

0.4- image pyramid or simple image scale
1 is number of pyramid layers. 1 mean that flow is calculated only from previous image. 
12 is win size.. Flow is computed over the window larger value is more robust to the noise. 
2 mean number of iteration of algorithm
8 is polynomial degree expansion recommended value are 5 - 7
1.2 standard deviation used to smooth used derivatives recommended values from 1.1 - 1,5

calcOpticalFlowFarneback(prevgray, img, flowUmat, 0.4, 1, 12, 2, 8, 1.2, 0);


calcOpticalFlowFarneback Example

Mine gopro video is captured at Old Town Square in Prague.. 






Opencv Tutorial C++ code


#include "opencv2\highgui.hpp"
#include "opencv2\imgproc.hpp"
#include "opencv2\objdetect\objdetect.hpp"
#include "opencv2/video/tracking.hpp"
#include <vector>
#include <stdio.h>
#include <Windows.h>
#include <iostream>


using namespace cv;
using namespace std;



int main(int argc, const char** argv)
{

// add your file name
VideoCapture cap("yourFile.mp4");

Mat flow, frame;
         // some faster than mat image container
UMat  flowUmat, prevgray;

for (;;)
{

bool Is = cap.grab();
if (Is == false) {
                         // if video capture failed
cout << "Video Capture Fail" << endl;
break;
}
else {

Mat img;
Mat original;

// capture frame from video file
cap.retrieve(img, CV_CAP_OPENNI_BGR_IMAGE);
resize(img, img, Size(640, 480));

// save original for later
img.copyTo(original);

// just make current frame gray
cvtColor(img, img, COLOR_BGR2GRAY);

// For all optical flow you need a sequence of images.. Or at least 2 of them. Previous                           //and current frame
//if there is no current frame
// go to this part and fill previous frame
//else {
// img.copyTo(prevgray);
//   }
// if previous frame is not empty.. There is a picture of previous frame. Do some                                  //optical flow alg. 

if (prevgray.empty() == false ) {

// calculate optical flow
calcOpticalFlowFarneback(prevgray, img, flowUmat, 0.4, 1, 12, 2, 8, 1.2, 0);
// copy Umat container to standard Mat
flowUmat.copyTo(flow);

                           
// By y += 5, x += 5 you can specify the grid 
for (int y = 0; y < original.rows; y += 5) {
for (int x = 0; x < original.cols; x += 5)
{
// get the flow from y, x position * 10 for better visibility
        const Point2f flowatxy = flow.at<Point2f>(y, x) * 10;
// draw line at flow direction
line(original, Point(x, y), Point(cvRound(x + flowatxy.x), cvRound(y + flowatxy.y)), Scalar(255,0,0));
                                                         // draw initial point
circle(original, Point(x, y), 1, Scalar(0, 0, 0), -1);

}

}
                                 // draw the results
namedWindow("prew", WINDOW_AUTOSIZE);
imshow("prew", original);
                                 // fill previous image again
img.copyTo(prevgray);

}
else {

                                 // fill previous image in case prevgray.empty() == true
img.copyTo(prevgray);
}

int key1 = waitKey(20);

}
}
}

23 comments:

  1. Hi Vladimir, I try to run your code, however it reported undeclared identifier: UMat. Is it only for GPU? Could you please inform how to run your code if I only have CPU? Thanks.

    ReplyDelete
    Replies
    1. Hello. :)

      I am just a guest.

      GMat is defined in the OpenCV 3.1.

      Delete
    2. Thank you Juhong! Is UMat only for GPU?

      Delete
    3. Umat is basically for OpenCL/ CPU whereas GpuMat is meant for allocating device memory in Nvidia-based GPUs.

      Delete
  2. can you please tell how to calculate velocity using optical flow

    ReplyDelete
    Replies
    1. yes find out the centroid and fps. do perspective projection and calculate speed

      Delete
    2. hello, can you share code for velocity calculation using optical flow....thanks

      Delete
  3. I do agree with all the ideas you have presented in your post. They’re really convincing and will certainly work. Still, the posts are very short for newbies. Could you please extend them a little from next time?..Keep this great work..
    GMAT Training in Chennai
    GMAT Coaching Chennai
    Best GMAT Coaching Classes in Chennai

    ReplyDelete

  4. Thanks for sharing the very useful info about clanguage and please keep updating........

    ReplyDelete
  5. Nice tutorial. Thanks for sharing the valuable info about c Training. it’s really helpful. Who want to learn c language this blog most helpful. Keep sharing on updated tutorials…..

    ReplyDelete
  6. Thanks for sharing the c++ tutorial with us and I have learned new programming skills especially on Optical Flow Farneback parameters remarks. I will bookmark this site so that I can practice writing the shared programs on my own especially when I have finish accessing Guide on Assignment Writing.

    ReplyDelete
  7. Can anybody give me same code of Optical Flow with Opencv2, Python Language

    ReplyDelete
  8. hey bro can you you help same project in java

    ReplyDelete
  9. This blog is awesome and very informative keep Sharing this type of blog.

    แตกใน xxx

    ReplyDelete
    Replies
    1. Thanks, i will keep sharing and please share my blog. That helps to doing this. :)

      Delete
  10. Thank you for such a nice detailed post. I always love to read your site content as these are really helpful for me.
    รับแทงบอล
    sbobet mobile
    royal1688
    ทางเข้า maxbet

    ReplyDelete
  11. If i give any video file as input to videoCapture then "cap.grab();" is returning false always. the code is working for vidoeCapture(0) (From webcam). need some help to debbug!!!

    ReplyDelete
  12. People tired to find a relevant place where they can know real facts and myths about the topic mentioned by author of this blog. I want to help a lot of needy people through this blog to come up at my blog to know the real facts and myths regarding this topic. computer vision apps

    ReplyDelete
  13. Hey Man its a nice tutorial but how to modify the code for occlusion culling?
    It would be nice if you share your thoughts!

    ReplyDelete
  14. This is very much interesting, thanks for sharing your ideas with us.
    DevOps Online Training

    ReplyDelete

ad