I would like to measure some coins placed on top of a calibration plate, as per this matlab example http://mathworks.com/help/vision/exam... .
Does OpenCV have an equivalent to the pointsToWorld function that lets you pass in pixel X,Y's and get world X,Y's on a plane?
There's no analog for pointsToWorld as I know. You can find x and y (z is equal zero in such case) like this:
cv::undistortPoints to your pixel coordintes (let's u and v) and get ideal coordinates (un and vn)solve system of linear equations:
w * [un vn 1] = Rt * [x y 0 1]
that gets you
r11 * x + r12 * y - w * un = -t1;
r21 * x + r22 * y - w * vn = -t2;
r31 * x + r32 * y - w = -t3;Here x and y are coordinates in chessboard plane, w is distant to point from camera, rij are components of rotation matrix and ti are components of translation vector.
Thanks, i'm still struggling to get the correct equation. My python to go from object points to image points looks like this:
object_points = np.array([[0, 0, .4], [.3, .2, .4]])
image_points, jacobian = cv2.projectPoints(object_points, rvec, tvec, camera_matrix, dist_coefs)
cv2.line(img, tuple(image_points[0].astype(int).ravel()), tuple(image_points[1].astype(int).ravel()),(0, 255, 0), 1)
image_points = np.array([[100, 100], [200, 200]])
object_points_on_plane = ????
but i'm unsure how to go back to object points on a plane from image points.
Asked: Jul 9 '16
Seen: 1,738 times
Last updated: Jul 22 '16
yes http://docs.opencv.org/3.1.0/d9/d0c/g... in opencv_contrib http://docs.opencv.org/3.1.0/d3/ddc/g...
thanks but I don't see anything in those links directly relating to the question. Am I missing something?