Computes the ideal point coordinates from the observed point coordinates

dst = cv.undistortPoints(src, cameraMatrix, distCoeffs)
dst = cv.undistortPoints(..., 'OptionName', optionValue, ...)

Input

• src Observed point coordinates. An Nx2, 1xNx2, or Nx1x2 array (either single or double).
• cameraMatrix Input camera matrix A = [fx 0 cx; 0 fy cy; 0 0 1].
• distCoeffs Input vector of distortion coefficients [k1,k2,p1,p2,k3,k4,k5,k6,s1,s2,s3,s4,taux,tauy] of 4, 5, 8, 12 or 14 elements. If the vector is empty, the zero distortion coefficients are assumed.

Output

• dst Output ideal point coordinates after undistortion and reverse perspective transformation. If matrix P is identity or omitted, dst will contain normalized point coordinates. Same size and type as the input points src.

Options

• R Rectification transformation in the object space (3x3 matrix). R1 or R2 computed by cv.stereoRectify can be passed here. If the matrix is empty, the identity transformation is used. default empty
• P New camera matrix (3x3) or new projection matrix (3x4). P1 or P2 computed by cv.stereoRectify can be passed here. If the matrix is empty, the identity new camera matrix is used. default empty

The function is similar to cv.undistort and cv.initUndistortRectifyMap but it operates on a sparse set of points instead of a raster image. Also the function performs a reverse transformation to cv.projectPoints. In case of a 3D object, it does not reconstruct its 3D coordinates, but for a planar object, it does, up to a translation vector, if the proper R is specified.

% (u,v) is the input point, (up, vp) is the output point
% camera_matrix = [fx 0 cx; 0 fy cy; 0 0 1]
% P = [fxp 0 cxp tx; 0 fyp cyp ty; 0 0 tz]
xpp = (u - cx)/fx
ypp = (v - cy)/fy
(xp,yp) = cv.undistort(xpp, ypp, dist_coeffs)
[X,Y,W]' = R*[xp yp 1]'
x = X/W, y = Y/W
up = x*fxp + cxp
vp = y*fyp + cyp

where cv.undistort is an approximate iterative algorithm that estimates the normalized original point coordinates out of the normalized distorted point coordinates ("normalized" means that the coordinates do not depend on the camera matrix).

The function can be used for both a stereo camera head or a monocular camera (when R is empty).