Background subtraction problem [closed]

asked 2013-03-03 03:22:03 -0600

hunter gravatar image

Hi I am working on a video processing project to detect foreground objects and my code to separate foreground and background is as follows :

#include "opencv2/core/core.hpp"
#include "opencv2/video/background_segm.hpp"
#include "opencv2/highgui/highgui.hpp"
#include "opencv/highgui.h"
#include <stdio.h>
#include <opencv2/imgproc/imgproc.hpp>


using namespace std;
using namespace cv;


//this is a sample for foreground detection functions
int main(int argc, const char** argv)
{


    VideoCapture cap;
    bool update_bg_model = true;

    cap.open(0);


    if( !cap.isOpened() )
    {
        printf("can not open camera or video file\n");
        return -1;
    }

    namedWindow("image", CV_WINDOW_NORMAL);
    namedWindow("foreground mask", CV_WINDOW_NORMAL);
    namedWindow("foreground image", CV_WINDOW_NORMAL);
    namedWindow("mean background image", CV_WINDOW_NORMAL);

    BackgroundSubtractorMOG2 bg_model(500,10,false);
    Mat img, fgmask, fgimg;

    for(;;)
    {
        cap >> img;

        if( img.empty() )
            break;

        if( fgimg.empty() )
          fgimg.create(img.size(), img.type());
        cv::Mat framecopy;


        //update the model
        blur(img,framecopy, Size(10,10), Point(-1,-1) );


        bg_model(framecopy, fgmask, update_bg_model ? -1 : 0);

        fgimg = Scalar::all(0);
        img.copyTo(fgimg, fgmask);

        Mat bgimg;

        bg_model.getBackgroundImage(bgimg);
        /*std::vector< std::vector<cv::Point> > contours;
        findContours(fgmask,contours,CV_RETR_EXTERNAL,CV_CHAIN_APPROX_SIMPLE,Point(0,0));
        std::vector<Rect> boundRect( contours.size() );

        for(int i=0;i<contours.size();i++)
        {
            if(contourArea(contours[i])>20)
            boundRect[i] = boundingRect( Mat(contours[i]) );
            cv::rectangle( img, boundRect[i].tl(), boundRect[i].br(), 255, 2, 8, 0 );
        }*/



        //erode(fgmask,fgmask,cv::Mat(50,50));

        imshow("image", img);
        imshow("foreground mask", fgmask);
        imshow("foreground image", fgimg);
        if(!bgimg.empty())
          imshow("mean background image", bgimg );

        char k = (char)waitKey(30);
        if( k == 27 ) break;
        if( k == ' ' )
        {
            update_bg_model = !update_bg_model;
            if(update_bg_model)
                printf("Background update is on\n");
            else
                printf("Background update is off\n");
        }
    }

    return 0;
}

and the resulting foregroundmask I got is : image description

t seems my code doesnt cover the whole object completely. Is there anyway I can improve it??

edit retag flag offensive reopen merge delete

Closed for the following reason question is not relevant or outdated by sturkmen
close date 2020-09-27 11:43:46.643196

Comments

hey hunter,

thank you so much your code helped me a lot. i am looking for the same code, i think you are using gaussian mixture model.

TJ gravatar imageTJ ( 2015-03-05 04:12:56 -0600 )edit

my project is summarize videos.

as my task is to detect foreground and background of frames in video.

with the help of your code this task was completed.

i am getting error

VideoCapture cap =cap.open("Sherlock outtakes.avi");
    bool update_bg_model = true;

instead of

VideoCapture cap;
    bool update_bg_model = true;

    cap.open(0);

but i am getting error "Unhandled exception at 0x00007FFAD94DCBE9 (opencv_highgui249d.dll) in dynamic.exe: 0xC0000005: Access violation reading location 0x0000000000000018" and the program breaks.

can you help me in this.

TJ gravatar imageTJ ( 2015-03-05 04:34:06 -0600 )edit