Is there a simple method to highlight the mask? - c++

Question

If I have mask like
And I have a image( the size is same to the mask) like
I want to hightlight the mask in the image. If I'm in other Language,I just
As you can see, the result image have a transparent red show the mask. I hope implement this in OpenCV. So I write this code
#include <opencv.hpp>
using namespace cv;
using namespace std;
int main() {
Mat srcImg = imread("image.jpg");
Mat mask = imread("mask.jpg", IMREAD_GRAYSCALE)>200;
for(int i=0;i<srcImg.rows;i++)
for(int j=0;j<srcImg.cols;j++)
if(mask.at<uchar>(i, j)==255)
circle(srcImg, Point(j,i), 3, Scalar(0, 0, 128,128));
imshow("image",srcImg);
waitKey();
return 0;
}
But as you see, I use a alpha value in Scalar, but it is not a transparent red.
Maybe this is due to the srcImg just have 3 channels. I have two question about this
How to hightlight the mask with a transparent red(even the image just have 3 channels)?
I have to draw circle pixel by pixel to do this thing?

Answer 1

#include<opencv2/core.hpp>
#include<opencv2/imgproc.hpp>
#include<opencv2/highgui.hpp>
using namespace cv;
int main(int argc, char** argv)
{
Mat srcImg = imread("image.png");
Mat mask = imread("mask.png", IMREAD_GRAYSCALE) > 200;
Mat red;
cvtColor(mask, red, COLOR_GRAY2BGR);
red = (red - Scalar(0, 0, 255)) / 2;
srcImg = srcImg - red;
imshow("image", srcImg);
waitKey();
return 0;
}

Answer 2

I've written this in python but you can easily port it to C++. Assuming that your source and mask images are CV_8UC3 images:
src = cv2.imread("source.png", -1)
mask = cv2.imread("mask.png", -1)
# convert mask to gray and then threshold it to convert it to binary
gray = cv2.cvtColor(mask, cv2.COLOR_BGR2GRAY)
ret, binary = cv2.threshold(gray, 40, 255, cv2.THRESH_BINARY)
# find contours of two major blobs present in the mask
im2,contours,hierarchy = cv2.findContours(binary, cv2.RETR_LIST, cv2.CHAIN_APPROX_NONE)
# draw the found contours on to source image
for contour in contours:
cv2.drawContours(src, contour, -1, (255,0,0), thickness = 1)
# split source to B,G,R channels
b,g,r = cv2.split(src)
# add a constant to R channel to highlight the selected area in reed
r = cv2.add(b, 30, dst = b, mask = binary, dtype = cv2.CV_8U)
# merge the channels back together
cv2.merge((b,g,r), src)