Problem with Watershed algorithm
How to find the contour of only the leg and foot parts?
Original image:
Denoised image:
Result:
Here is my code:
#include "opencv2/imgproc/imgproc.hpp"
#include "opencv2/highgui/highgui.hpp"
#include "opencv2/opencv.hpp"
#include <string>
using namespace cv;
using namespace std;
class WatershedSegmenter {
private:
cv::Mat markers;
public:
void setMarkers(cv::Mat& markerImage)
{
markerImage.convertTo(markers, CV_32S);
}
cv::Mat process(cv::Mat &image)
{
cv::watershed(image, markers);
markers.convertTo(markers, CV_8U);
return markers;
}
};
int main(int argc, char* argv[])
{
cv::Mat image = cv::imread("C:/Proj/Denoised.jpg");
cv::Mat blank(image.size(), CV_8U, cv::Scalar(0xFF));
cv::Mat dest;
imshow("originalimage", image);
// Create markers image
cv::Mat markers(image.size(), CV_8U, cv::Scalar(-1));
//Rect(topleftcornerX, topleftcornerY, width, height);
//top rectangle
markers(Rect(0, 0, image.cols, 5)) = Scalar::all(1);
//bottom rectangle
markers(Rect(0, image.rows - 5, image.cols, 5)) = Scalar::all(1);
//left rectangle
markers(Rect(0, 0, 5, image.rows)) = Scalar::all(1);
//right rectangle
markers(Rect(image.cols - 5, 0, 5, image.rows)) = Scalar::all(1);
//centre rectangle
int centreW = image.cols / 4;
int centreH = image.rows / 4;
markers(Rect((image.cols / 2) - (centreW / 2), (image.rows / 2) - (centreH / 2), centreW, centreH)) = Scalar::all(2);
markers.convertTo(markers, CV_BGR2GRAY);
imshow("markers", markers);
//Create watershed segmentation object
WatershedSegmenter segmenter;
segmenter.setMarkers(markers);
cv::Mat wshedMask = segmenter.process(image);
cv::Mat mask;
convertScaleAbs(wshedMask, mask, 1, 0);
double thresh = threshold(mask, mask, 1, 255, THRESH_BINARY);
bitwise_and(image, image, dest, mask);
dest.convertTo(dest, CV_8U);
imshow("final_result", dest);
//save img
imwrite("C:/Proj/watershed.jpg", dest);
cv::waitKey(0);
return 0;
}
take a look at this you can try to detect skin
I don't think it is possible with watershed. With grabcut it is possible if you choose mask carefully