background subtraction
I was using and testing MOG2 background subtraction and othres provided by opencv. But it looks like thsi function don't takes as still image as a background instead takes it every time. I want to use one single image as a background and than use forground image to be subracted and give the result. Isn't it possible using any opencv function? This is the code i was using to check the mog2 function:
#include <iostream>
#include <sstream>
#include <opencv2/imgcodecs.hpp>
#include <opencv2/imgproc.hpp>
#include <opencv2/videoio.hpp>
#include <opencv2/highgui.hpp>
#include <opencv2/video.hpp>
using namespace cv;
using namespace std;
const char* params
= "{ help h | | Print usage }"
"{ input | vtest.avi | Path to a video or a sequence of image }"
"{ algo | MOG2 | Background subtraction method (KNN, MOG2) }";
int main(int argc, char* argv[])
{
CommandLineParser parser(argc, argv, params);
parser.about( "This program shows how to use background subtraction methods provided by "
" OpenCV. You can process both videos and images.\n" );
if (parser.has("help"))
{
//print help information
parser.printMessage();
}
//create Background Subtractor objects
Ptr<BackgroundSubtractor> pBackSub;
/*if (parser.get<String>("algo") == "MOG2")
pBackSub = createBackgroundSubtractorMOG2();
else
pBackSub = createBackgroundSubtractorKNN();*/
pBackSub = createBackgroundSubtractorMOG2();
VideoCapture cap;
if(!cap.open(0))
return 0;
/*VideoCapture capture( samples::findFile( parser.get<String>("input") ) );
if (!capture.isOpened()){
//error in opening the video input
cerr << "Unable to open: " << parser.get<String>("input") << endl;
return 0;
}*/
Mat frame, fgMask;
while (true) {
cap >> frame;
if (frame.empty())
break;
//update the background model
pBackSub->apply(frame, fgMask);
//get the frame number and write it on the current frame
rectangle(frame, cv::Point(10, 2), cv::Point(100,20),
cv::Scalar(255,255,255), -1);
stringstream ss;
ss << cap.get(CAP_PROP_POS_FRAMES);
string frameNumberString = ss.str();
putText(frame, frameNumberString.c_str(), cv::Point(15, 15),
FONT_HERSHEY_SIMPLEX, 0.5 , cv::Scalar(0,0,0));
//show the current frame and the fg masks
imshow("Frame", frame);
imshow("FG Mask", fgMask);
//get the input from the keyboard
int keyboard = waitKey(30);
if (keyboard == 'q' || keyboard == 27)
break;
}
return 0;
}
why do you try to use MOG2 then,where a simple
absdiff()
would do ?what are you trying to achieve ? what is the purpose of your program ?
What i want is i have a camera taking a background image stream contineuly. Now when new object enters in this scene its pixels and controus should be extracted till it is inside the camera scene. as i used mog2 after few seconds the newly entered object is also considered as background in it and also the contour and the body pixel is not that good. Now, i tried as u suggested absdiff() and did it myself using different color models like lab, luv, hsv etc, but i found that due to noise and other factors the result is also little bit of noisy. i tried noise removal functions like fastNlMeansDenoisingColoredMulti but it is making the frame rate drop drastically where the mog2 has good frame rate without having such noise.