I have a road scene picture, and I can detect the vehicles on the road. What I want to do is to blur the whole picture except the vehicle detection areas, my solution is firstly I create a empty Mat and crop the vehicle detection area into the empty Mat, and secondly blur the whole picture, now the problem is how to combine the blury picture with the vehicle parts ?
You can just copy the vehicles using cv::copy. If A is your full (unblurred Image), B is the blurred version, and a car region is defined from col c_1 to c_2 and row range r_1,r_2, you can copy the car by
A.rowrange(r_1,r_2).colrange(c_1,c_2).copyTo(B.rowrange(r_1,r_2).colrange(c_1,c_2))
Hi!
This can be done using additions (easy to code, but not very efficient...):
Mat imgSrc;//here is your input img
vector<Rect> boundingBoxes;//and the bounding boxes
Mat imgBlur, imgMask;
blur( imgSrc, imgBlur, Size( 5, 5 ) );
imgMask = Mat::zeros(imgSrc.size(), CV_8UC1);
for(size_t i=0; i<boundingBoxes.size(); i++)
{
//Remove part of image
rectangle(imgBlur, boundingBoxes[i], Scalar(0), CV_FILLED);
//and create the mask
rectangle(imgMask, boundingBoxes[i], Scalar(255), CV_FILLED);
}
Mat output = imgBlur.clone();
add(imgBlur, imgSrc, output, imgMask);
1-You have localize the external mask of all vehicles.
2-Apply radial low pass filter(in the edge,the sigma is desired value & in the center of mask sigma is zero) on mask region of each vehicles.
3-Apply low pass filter on the whole image apart from vehicles mask region.
4-Copy the vehicles region from the radial low pass image in to destination low pass image.
Asked: 2014-11-29 01:30:19 -0600
Seen: 912 times
Last updated: Dec 01 '14
