OpenCV  4.1.0
Open Source Computer Vision
Namespaces | Enumerations | Functions

Namespaces

 cv::fisheye
 The methods in this namespace use a so-called fisheye camera model.
 

Enumerations

enum  {
  cv::fisheye::CALIB_USE_INTRINSIC_GUESS = 1 << 0,
  cv::fisheye::CALIB_RECOMPUTE_EXTRINSIC = 1 << 1,
  cv::fisheye::CALIB_CHECK_COND = 1 << 2,
  cv::fisheye::CALIB_FIX_SKEW = 1 << 3,
  cv::fisheye::CALIB_FIX_K1 = 1 << 4,
  cv::fisheye::CALIB_FIX_K2 = 1 << 5,
  cv::fisheye::CALIB_FIX_K3 = 1 << 6,
  cv::fisheye::CALIB_FIX_K4 = 1 << 7,
  cv::fisheye::CALIB_FIX_INTRINSIC = 1 << 8,
  cv::fisheye::CALIB_FIX_PRINCIPAL_POINT = 1 << 9
}
 

Functions

double cv::fisheye::calibrate (InputArrayOfArrays objectPoints, InputArrayOfArrays imagePoints, const Size &image_size, InputOutputArray K, InputOutputArray D, OutputArrayOfArrays rvecs, OutputArrayOfArrays tvecs, int flags=0, TermCriteria criteria=TermCriteria(TermCriteria::COUNT+TermCriteria::EPS, 100, DBL_EPSILON))
 Performs camera calibaration. More...
 
void cv::fisheye::distortPoints (InputArray undistorted, OutputArray distorted, InputArray K, InputArray D, double alpha=0)
 Distorts 2D points using fisheye model. More...
 
void cv::fisheye::estimateNewCameraMatrixForUndistortRectify (InputArray K, InputArray D, const Size &image_size, InputArray R, OutputArray P, double balance=0.0, const Size &new_size=Size(), double fov_scale=1.0)
 Estimates new camera matrix for undistortion or rectification. More...
 
void cv::fisheye::initUndistortRectifyMap (InputArray K, InputArray D, InputArray R, InputArray P, const cv::Size &size, int m1type, OutputArray map1, OutputArray map2)
 Computes undistortion and rectification maps for image transform by cv::remap(). If D is empty zero distortion is used, if R or P is empty identity matrixes are used. More...
 
void cv::fisheye::projectPoints (InputArray objectPoints, OutputArray imagePoints, const Affine3d &affine, InputArray K, InputArray D, double alpha=0, OutputArray jacobian=noArray())
 Projects points using fisheye model. More...
 
void cv::fisheye::projectPoints (InputArray objectPoints, OutputArray imagePoints, InputArray rvec, InputArray tvec, InputArray K, InputArray D, double alpha=0, OutputArray jacobian=noArray())
 
double cv::fisheye::stereoCalibrate (InputArrayOfArrays objectPoints, InputArrayOfArrays imagePoints1, InputArrayOfArrays imagePoints2, InputOutputArray K1, InputOutputArray D1, InputOutputArray K2, InputOutputArray D2, Size imageSize, OutputArray R, OutputArray T, int flags=fisheye::CALIB_FIX_INTRINSIC, TermCriteria criteria=TermCriteria(TermCriteria::COUNT+TermCriteria::EPS, 100, DBL_EPSILON))
 Performs stereo calibration. More...
 
void cv::fisheye::stereoRectify (InputArray K1, InputArray D1, InputArray K2, InputArray D2, const Size &imageSize, InputArray R, InputArray tvec, OutputArray R1, OutputArray R2, OutputArray P1, OutputArray P2, OutputArray Q, int flags, const Size &newImageSize=Size(), double balance=0.0, double fov_scale=1.0)
 Stereo rectification for fisheye camera model. More...
 
void cv::fisheye::undistortImage (InputArray distorted, OutputArray undistorted, InputArray K, InputArray D, InputArray Knew=cv::noArray(), const Size &new_size=Size())
 Transforms an image to compensate for fisheye lens distortion. More...
 
void cv::fisheye::undistortPoints (InputArray distorted, OutputArray undistorted, InputArray K, InputArray D, InputArray R=noArray(), InputArray P=noArray())
 Undistorts 2D points using fisheye model. More...
 

Detailed Description

Definitions: Let P be a point in 3D of coordinates X in the world reference frame (stored in the matrix X) The coordinate vector of P in the camera reference frame is:

\[Xc = R X + T\]

where R is the rotation matrix corresponding to the rotation vector om: R = rodrigues(om); call x, y and z the 3 coordinates of Xc:

\[x = Xc_1 \\ y = Xc_2 \\ z = Xc_3\]

The pinhole projection coordinates of P is [a; b] where

\[a = x / z \ and \ b = y / z \\ r^2 = a^2 + b^2 \\ \theta = atan(r)\]

Fisheye distortion:

\[\theta_d = \theta (1 + k_1 \theta^2 + k_2 \theta^4 + k_3 \theta^6 + k_4 \theta^8)\]

The distorted point coordinates are [x'; y'] where

\[x' = (\theta_d / r) a \\ y' = (\theta_d / r) b \]

Finally, conversion into pixel coordinates: The final pixel coordinates vector [u; v] where:

\[u = f_x (x' + \alpha y') + c_x \\ v = f_y y' + c_y\]

Enumeration Type Documentation

§ anonymous enum

anonymous enum

#include <opencv2/calib3d.hpp>

Enumerator
CALIB_USE_INTRINSIC_GUESS 
Python: cv.fisheye.CALIB_USE_INTRINSIC_GUESS
CALIB_RECOMPUTE_EXTRINSIC 
Python: cv.fisheye.CALIB_RECOMPUTE_EXTRINSIC
CALIB_CHECK_COND 
Python: cv.fisheye.CALIB_CHECK_COND
CALIB_FIX_SKEW 
Python: cv.fisheye.CALIB_FIX_SKEW
CALIB_FIX_K1 
Python: cv.fisheye.CALIB_FIX_K1
CALIB_FIX_K2 
Python: cv.fisheye.CALIB_FIX_K2
CALIB_FIX_K3 
Python: cv.fisheye.CALIB_FIX_K3
CALIB_FIX_K4 
Python: cv.fisheye.CALIB_FIX_K4
CALIB_FIX_INTRINSIC 
Python: cv.fisheye.CALIB_FIX_INTRINSIC
CALIB_FIX_PRINCIPAL_POINT 
Python: cv.fisheye.CALIB_FIX_PRINCIPAL_POINT

Function Documentation

§ calibrate()

double cv::fisheye::calibrate ( InputArrayOfArrays  objectPoints,
InputArrayOfArrays  imagePoints,
const Size image_size,
InputOutputArray  K,
InputOutputArray  D,
OutputArrayOfArrays  rvecs,
OutputArrayOfArrays  tvecs,
int  flags = 0,
TermCriteria  criteria = TermCriteria(TermCriteria::COUNT+TermCriteria::EPS, 100, DBL_EPSILON) 
)
Python:
retval, K, D, rvecs, tvecs=cv.fisheye.calibrate(objectPoints, imagePoints, image_size, K, D[, rvecs[, tvecs[, flags[, criteria]]]])

#include <opencv2/calib3d.hpp>

Performs camera calibaration.

Parameters
objectPointsvector of vectors of calibration pattern points in the calibration pattern coordinate space.
imagePointsvector of vectors of the projections of calibration pattern points. imagePoints.size() and objectPoints.size() and imagePoints[i].size() must be equal to objectPoints[i].size() for each i.
image_sizeSize of the image used only to initialize the intrinsic camera matrix.
KOutput 3x3 floating-point camera matrix \(A = \vecthreethree{f_x}{0}{c_x}{0}{f_y}{c_y}{0}{0}{1}\) . If fisheye::CALIB_USE_INTRINSIC_GUESS/ is specified, some or all of fx, fy, cx, cy must be initialized before calling the function.
DOutput vector of distortion coefficients \((k_1, k_2, k_3, k_4)\).
rvecsOutput vector of rotation vectors (see Rodrigues ) estimated for each pattern view. That is, each k-th rotation vector together with the corresponding k-th translation vector (see the next output parameter description) brings the calibration pattern from the model coordinate space (in which object points are specified) to the world coordinate space, that is, a real position of the calibration pattern in the k-th pattern view (k=0.. M -1).
tvecsOutput vector of translation vectors estimated for each pattern view.
flagsDifferent flags that may be zero or a combination of the following values:
criteriaTermination criteria for the iterative optimization algorithm.

§ distortPoints()

void cv::fisheye::distortPoints ( InputArray  undistorted,
OutputArray  distorted,
InputArray  K,
InputArray  D,
double  alpha = 0 
)
Python:
distorted=cv.fisheye.distortPoints(undistorted, K, D[, distorted[, alpha]])

#include <opencv2/calib3d.hpp>

Distorts 2D points using fisheye model.

Parameters
undistortedArray of object points, 1xN/Nx1 2-channel (or vector<Point2f> ), where N is the number of points in the view.
KCamera matrix \(K = \vecthreethree{f_x}{0}{c_x}{0}{f_y}{c_y}{0}{0}{_1}\).
DInput vector of distortion coefficients \((k_1, k_2, k_3, k_4)\).
alphaThe skew coefficient.
distortedOutput array of image points, 1xN/Nx1 2-channel, or vector<Point2f> .

Note that the function assumes the camera matrix of the undistorted points to be identity. This means if you want to transform back points undistorted with undistortPoints() you have to multiply them with \(P^{-1}\).

§ estimateNewCameraMatrixForUndistortRectify()

void cv::fisheye::estimateNewCameraMatrixForUndistortRectify ( InputArray  K,
InputArray  D,
const Size image_size,
InputArray  R,
OutputArray  P,
double  balance = 0.0,
const Size new_size = Size(),
double  fov_scale = 1.0 
)
Python:
P=cv.fisheye.estimateNewCameraMatrixForUndistortRectify(K, D, image_size, R[, P[, balance[, new_size[, fov_scale]]]])

#include <opencv2/calib3d.hpp>

Estimates new camera matrix for undistortion or rectification.

Parameters
KCamera matrix \(K = \vecthreethree{f_x}{0}{c_x}{0}{f_y}{c_y}{0}{0}{_1}\).
image_size
DInput vector of distortion coefficients \((k_1, k_2, k_3, k_4)\).
RRectification transformation in the object space: 3x3 1-channel, or vector: 3x1/1x3 1-channel or 1x1 3-channel
PNew camera matrix (3x3) or new projection matrix (3x4)
balanceSets the new focal length in range between the min focal length and the max focal length. Balance is in range of [0, 1].
new_size
fov_scaleDivisor for new focal length.

§ initUndistortRectifyMap()

void cv::fisheye::initUndistortRectifyMap ( InputArray  K,
InputArray  D,
InputArray  R,
InputArray  P,
const cv::Size size,
int  m1type,
OutputArray  map1,
OutputArray  map2 
)
Python:
map1, map2=cv.fisheye.initUndistortRectifyMap(K, D, R, P, size, m1type[, map1[, map2]])

#include <opencv2/calib3d.hpp>

Computes undistortion and rectification maps for image transform by cv::remap(). If D is empty zero distortion is used, if R or P is empty identity matrixes are used.

Parameters
KCamera matrix \(K = \vecthreethree{f_x}{0}{c_x}{0}{f_y}{c_y}{0}{0}{_1}\).
DInput vector of distortion coefficients \((k_1, k_2, k_3, k_4)\).
RRectification transformation in the object space: 3x3 1-channel, or vector: 3x1/1x3 1-channel or 1x1 3-channel
PNew camera matrix (3x3) or new projection matrix (3x4)
sizeUndistorted image size.
m1typeType of the first output map that can be CV_32FC1 or CV_16SC2 . See convertMaps() for details.
map1The first output map.
map2The second output map.

§ projectPoints() [1/2]

void cv::fisheye::projectPoints ( InputArray  objectPoints,
OutputArray  imagePoints,
const Affine3d affine,
InputArray  K,
InputArray  D,
double  alpha = 0,
OutputArray  jacobian = noArray() 
)
Python:
imagePoints, jacobian=cv.fisheye.projectPoints(objectPoints, rvec, tvec, K, D[, imagePoints[, alpha[, jacobian]]])

#include <opencv2/calib3d.hpp>

Projects points using fisheye model.

Parameters
objectPointsArray of object points, 1xN/Nx1 3-channel (or vector<Point3f> ), where N is the number of points in the view.
imagePointsOutput array of image points, 2xN/Nx2 1-channel or 1xN/Nx1 2-channel, or vector<Point2f>.
affine
KCamera matrix \(K = \vecthreethree{f_x}{0}{c_x}{0}{f_y}{c_y}{0}{0}{_1}\).
DInput vector of distortion coefficients \((k_1, k_2, k_3, k_4)\).
alphaThe skew coefficient.
jacobianOptional output 2Nx15 jacobian matrix of derivatives of image points with respect to components of the focal lengths, coordinates of the principal point, distortion coefficients, rotation vector, translation vector, and the skew. In the old interface different components of the jacobian are returned via different output parameters.

The function computes projections of 3D points to the image plane given intrinsic and extrinsic camera parameters. Optionally, the function computes Jacobians - matrices of partial derivatives of image points coordinates (as functions of all the input parameters) with respect to the particular parameters, intrinsic and/or extrinsic.

§ projectPoints() [2/2]

void cv::fisheye::projectPoints ( InputArray  objectPoints,
OutputArray  imagePoints,
InputArray  rvec,
InputArray  tvec,
InputArray  K,
InputArray  D,
double  alpha = 0,
OutputArray  jacobian = noArray() 
)
Python:
imagePoints, jacobian=cv.fisheye.projectPoints(objectPoints, rvec, tvec, K, D[, imagePoints[, alpha[, jacobian]]])

#include <opencv2/calib3d.hpp>

This is an overloaded member function, provided for convenience. It differs from the above function only in what argument(s) it accepts.

§ stereoCalibrate()

double cv::fisheye::stereoCalibrate ( InputArrayOfArrays  objectPoints,
InputArrayOfArrays  imagePoints1,
InputArrayOfArrays  imagePoints2,
InputOutputArray  K1,
InputOutputArray  D1,
InputOutputArray  K2,
InputOutputArray  D2,
Size  imageSize,
OutputArray  R,
OutputArray  T,
int  flags = fisheye::CALIB_FIX_INTRINSIC,
TermCriteria  criteria = TermCriteria(TermCriteria::COUNT+TermCriteria::EPS, 100, DBL_EPSILON) 
)
Python:
retval, K1, D1, K2, D2, R, T=cv.fisheye.stereoCalibrate(objectPoints, imagePoints1, imagePoints2, K1, D1, K2, D2, imageSize[, R[, T[, flags[, criteria]]]])

#include <opencv2/calib3d.hpp>

Performs stereo calibration.

Parameters
objectPointsVector of vectors of the calibration pattern points.
imagePoints1Vector of vectors of the projections of the calibration pattern points, observed by the first camera.
imagePoints2Vector of vectors of the projections of the calibration pattern points, observed by the second camera.
K1Input/output first camera matrix: \(\vecthreethree{f_x^{(j)}}{0}{c_x^{(j)}}{0}{f_y^{(j)}}{c_y^{(j)}}{0}{0}{1}\) , \(j = 0,\, 1\) . If any of fisheye::CALIB_USE_INTRINSIC_GUESS , fisheye::CALIB_FIX_INTRINSIC are specified, some or all of the matrix components must be initialized.
D1Input/output vector of distortion coefficients \((k_1, k_2, k_3, k_4)\) of 4 elements.
K2Input/output second camera matrix. The parameter is similar to K1 .
D2Input/output lens distortion coefficients for the second camera. The parameter is similar to D1 .
imageSizeSize of the image used only to initialize intrinsic camera matrix.
ROutput rotation matrix between the 1st and the 2nd camera coordinate systems.
TOutput translation vector between the coordinate systems of the cameras.
flagsDifferent flags that may be zero or a combination of the following values:
criteriaTermination criteria for the iterative optimization algorithm.

§ stereoRectify()

void cv::fisheye::stereoRectify ( InputArray  K1,
InputArray  D1,
InputArray  K2,
InputArray  D2,
const Size imageSize,
InputArray  R,
InputArray  tvec,
OutputArray  R1,
OutputArray  R2,
OutputArray  P1,
OutputArray  P2,
OutputArray  Q,
int  flags,
const Size newImageSize = Size(),
double  balance = 0.0,
double  fov_scale = 1.0 
)
Python:
R1, R2, P1, P2, Q=cv.fisheye.stereoRectify(K1, D1, K2, D2, imageSize, R, tvec, flags[, R1[, R2[, P1[, P2[, Q[, newImageSize[, balance[, fov_scale]]]]]]]])

#include <opencv2/calib3d.hpp>

Stereo rectification for fisheye camera model.

Parameters
K1First camera matrix.
D1First camera distortion parameters.
K2Second camera matrix.
D2Second camera distortion parameters.
imageSizeSize of the image used for stereo calibration.
RRotation matrix between the coordinate systems of the first and the second cameras.
tvecTranslation vector between coordinate systems of the cameras.
R1Output 3x3 rectification transform (rotation matrix) for the first camera.
R2Output 3x3 rectification transform (rotation matrix) for the second camera.
P1Output 3x4 projection matrix in the new (rectified) coordinate systems for the first camera.
P2Output 3x4 projection matrix in the new (rectified) coordinate systems for the second camera.
QOutput \(4 \times 4\) disparity-to-depth mapping matrix (see reprojectImageTo3D ).
flagsOperation flags that may be zero or CALIB_ZERO_DISPARITY . If the flag is set, the function makes the principal points of each camera have the same pixel coordinates in the rectified views. And if the flag is not set, the function may still shift the images in the horizontal or vertical direction (depending on the orientation of epipolar lines) to maximize the useful image area.
newImageSizeNew image resolution after rectification. The same size should be passed to initUndistortRectifyMap (see the stereo_calib.cpp sample in OpenCV samples directory). When (0,0) is passed (default), it is set to the original imageSize . Setting it to larger value can help you preserve details in the original image, especially when there is a big radial distortion.
balanceSets the new focal length in range between the min focal length and the max focal length. Balance is in range of [0, 1].
fov_scaleDivisor for new focal length.

§ undistortImage()

void cv::fisheye::undistortImage ( InputArray  distorted,
OutputArray  undistorted,
InputArray  K,
InputArray  D,
InputArray  Knew = cv::noArray(),
const Size new_size = Size() 
)
Python:
undistorted=cv.fisheye.undistortImage(distorted, K, D[, undistorted[, Knew[, new_size]]])

#include <opencv2/calib3d.hpp>

Transforms an image to compensate for fisheye lens distortion.

Parameters
distortedimage with fisheye lens distortion.
undistortedOutput image with compensated fisheye lens distortion.
KCamera matrix \(K = \vecthreethree{f_x}{0}{c_x}{0}{f_y}{c_y}{0}{0}{_1}\).
DInput vector of distortion coefficients \((k_1, k_2, k_3, k_4)\).
KnewCamera matrix of the distorted image. By default, it is the identity matrix but you may additionally scale and shift the result by using a different matrix.
new_sizeThe function transforms an image to compensate radial and tangential lens distortion.

The function is simply a combination of fisheye::initUndistortRectifyMap (with unity R ) and remap (with bilinear interpolation). See the former function for details of the transformation being performed.

See below the results of undistortImage.

  • a) result of undistort of perspective camera model (all possible coefficients (k_1, k_2, k_3, k_4, k_5, k_6) of distortion were optimized under calibration)
    • b) result of fisheye::undistortImage of fisheye camera model (all possible coefficients (k_1, k_2, k_3, k_4) of fisheye distortion were optimized under calibration)
    • c) original image was captured with fisheye lens

Pictures a) and b) almost the same. But if we consider points of image located far from the center of image, we can notice that on image a) these points are distorted.

fisheye_undistorted.jpg
image

§ undistortPoints()

void cv::fisheye::undistortPoints ( InputArray  distorted,
OutputArray  undistorted,
InputArray  K,
InputArray  D,
InputArray  R = noArray(),
InputArray  P = noArray() 
)
Python:
undistorted=cv.fisheye.undistortPoints(distorted, K, D[, undistorted[, R[, P]]])

#include <opencv2/calib3d.hpp>

Undistorts 2D points using fisheye model.

Parameters
distortedArray of object points, 1xN/Nx1 2-channel (or vector<Point2f> ), where N is the number of points in the view.
KCamera matrix \(K = \vecthreethree{f_x}{0}{c_x}{0}{f_y}{c_y}{0}{0}{_1}\).
DInput vector of distortion coefficients \((k_1, k_2, k_3, k_4)\).
RRectification transformation in the object space: 3x3 1-channel, or vector: 3x1/1x3 1-channel or 1x1 3-channel
PNew camera matrix (3x3) or new projection matrix (3x4)
undistortedOutput array of image points, 1xN/Nx1 2-channel, or vector<Point2f> .