0% found this document useful (0 votes)
176 views5 pages

What Is Camera Calibration - MATLAB & Simulink

Camera calibration estimates camera parameters to correct lens distortion, measure object size, and determine camera location. It requires 3D world points and corresponding 2D image points. Parameters include intrinsics like focal length and extrinsics representing camera position. Lens distortion is also modeled.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
176 views5 pages

What Is Camera Calibration - MATLAB & Simulink

Camera calibration estimates camera parameters to correct lens distortion, measure object size, and determine camera location. It requires 3D world points and corresponding 2D image points. Parameters include intrinsics like focal length and extrinsics representing camera position. Lens distortion is also modeled.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

8/11/2016

WhatIsCameraCalibration?MATLAB&Simulink

WhatIsCameraCalibration?
Geometriccameracalibration,alsoreferredtoascameraresectioning,estimatestheparametersofalensandimage
sensorofanimageorvideocamera.Youcanusetheseparameterstocorrectforlensdistortion,measurethesizeof
anobjectinworldunits,ordeterminethelocationofthecamerainthescene.Thesetasksareusedinapplications
suchasmachinevisiontodetectandmeasureobjects.Theyarealsousedinrobotics,fornavigationsystems,and3
Dscenereconstruction.
Examplesofwhatyoucandoaftercalibratingyourcamera:

Cameraparametersincludeintrinsics,extrinsics,anddistortioncoefficients.Toestimatethecameraparameters,you
needtohave3Dworldpointsandtheircorresponding2Dimagepoints.Youcangetthesecorrespondencesusing
multipleimagesofacalibrationpattern,suchasacheckerboard.Usingthecorrespondences,youcansolveforthe
cameraparameters.Afteryoucalibrateacamera,toevaluatetheaccuracyoftheestimatedparameters,youcan:

Plottherelativelocationsofthecameraandthecalibrationpattern

Calculatethereprojectionerrors.

Calculatetheparameterestimationerrors.

UsetheCameraCalibratortoperformcameracalibrationandevaluatetheaccuracyoftheestimatedparameters.

CameraModel
TheComputerVisionSystemToolboxcalibrationalgorithmusesthecameramodelproposedbyJeanYvesBouguet
[3].Themodelincludes:

Thepinholecameramodel[1].

Lensdistortion[2].

Thepinholecameramodeldoesnotaccountforlensdistortionbecauseanidealpinholecameradoesnothavealens.
Toaccuratelyrepresentarealcamera,thefullcameramodelusedbythealgorithmincludestheradialandtangential
lensdistortion.

PinholeCameraModel
Apinholecameraisasimplecamerawithoutalensandwithasinglesmallaperture.Lightrayspassthroughthe
apertureandprojectaninvertedimageontheoppositesideofthecamera.Thinkofthevirtualimageplaneasbeingin
frontofthecameraandcontainingtheuprightimageofthescene.

http://www.mathworks.com/help/vision/ug/cameracalibration.html?requestedDomain=www.mathworks.com

1/5

8/11/2016

WhatIsCameraCalibration?MATLAB&Simulink

Thepinholecameraparametersarerepresentedina4by3matrixcalledthecameramatrix.Thismatrixmapsthe3D
worldsceneintotheimageplane.Thecalibrationalgorithmcalculatesthecameramatrixusingtheextrinsicand
intrinsicparameters.Theextrinsicparametersrepresentthelocationofthecamerainthe3Dscene.Theintrinsic
parametersrepresenttheopticalcenterandfocallengthofthecamera.

Theworldpointsaretransformedtocameracoordinatesusingtheextrinsicsparameters.Thecameracoordinatesare
mappedintotheimageplaneusingtheintrinsicsparameters.

CameraCalibrationParameters
Thecalibrationalgorithmcalculatesthecameramatrixusingtheextrinsicandintrinsicparameters.Theextrinsic
parametersrepresentarigidtransformationfrom3Dworldcoordinatesystemtothe3Dcamera'scoordinatesystem.
Theintrinsicparametersrepresentaprojectivetransformationfromthe3Dcamera'scoordinatesintothe2Dimage
coordinates.

ExtrinsicParameters
Theextrinsicparametersconsistofarotation,R,andatranslation,t.Theoriginofthecamera'scoordinatesystemis
atitsopticalcenteranditsxandyaxisdefinetheimageplane.
http://www.mathworks.com/help/vision/ug/cameracalibration.html?requestedDomain=www.mathworks.com

2/5

8/11/2016

WhatIsCameraCalibration?MATLAB&Simulink

IntrinsicParameters
Theintrinsicparametersincludethefocallength,theopticalcenter,alsoknownastheprincipalpoint,andtheskew
coefficient.Thecameraintrinsicmatrix,K,isdefinedas:

fx 0 0

s fy 0

cx cy 1
Thepixelskewisdefinedas:

[cx cy] Opticalcenter(theprincipalpoint),inpixels.


(fx, fy) Focallengthinpixels.
fx = F /px
fy = F /py
F Focallengthinworldunits,typicallyexpressedinmillimeters.
(px, py) Sizeofthepixelinworldunits.
s Skewcoefficient,whichisnonzeroiftheimageaxesarenotperpendicular.
s = fy tan
DistortioninCameraCalibration
Thecameramatrixdoesnotaccountforlensdistortionbecauseanidealpinholecameradoesnothavealens.To
accuratelyrepresentarealcamera,thecameramodelincludestheradialandtangentiallensdistortion.
RadialDistortion
Radialdistortionoccurswhenlightraysbendmoreneartheedgesofalensthantheydoatitsopticalcenter.The
smallerthelens,thegreaterthedistortion.

Theradialdistortioncoefficientsmodelthistypeofdistortion.Thedistortedpointsaredenotedas(x distorted,y distorted):


xdistorted=x(1+k1*r2+k2*r4+k3*r6)
2

http://www.mathworks.com/help/vision/ug/cameracalibration.html?requestedDomain=www.mathworks.com

3/5

8/11/2016

WhatIsCameraCalibration?MATLAB&Simulink
2

ydistorted=y(1+k1*r +k2*r +k3*r )

x,yUndistortedpixellocations.xandyareinnormalizedimagecoordinates.Normalizedimagecoordinatesare
calculatedfrompixelcoordinatesbytranslatingtotheopticalcenteranddividingbythefocallengthinpixels.
Thus,xandyaredimensionless.

k 1,k 2,andk 3Radialdistortioncoefficientsofthelens.

r2:x 2+y 2

Typically,twocoefficientsaresufficientforcalibration.Forseveredistortion,suchasinwideanglelenses,youcan
select3coefficientstoincludek 3.
TangentialDistortion
Tangentialdistortionoccurswhenthelensandtheimageplanearenotparallel.Thetangentialdistortioncoefficients
modelthistypeofdistortion.

Thedistortedpointsaredenotedas(x distorted,y distorted):


xdistorted=x+[2*p1*x*y+p2*(r2+2*x2)]
ydistorted=y+[p1*(r2+2*y2)+2*p2*x*y]

x,yUndistortedpixellocations.xandyareinnormalizedimagecoordinates.Normalizedimagecoordinatesare
calculatedfrompixelcoordinatesbytranslatingtotheopticalcenteranddividingbythefocallengthinpixels.
Thus,xandyaredimensionless.

p1andp2Tangentialdistortioncoefficientsofthelens.

r2=x 2+y 2

References
[1]Zhang,Z."AFlexibleNewTechniqueforCameraCalibration."IEEETransactionsonPatternAnalysisandMachine
Intelligence.Vol.22,No.11,2000,pp.13301334.
[2]Heikkila,J.,andO.Silven."AFourstepCameraCalibrationProcedurewithImplicitImageCorrection."IEEE
InternationalConferenceonComputerVisionandPatternRecognition.1997.
[3]Bouguet,J.Y."CameraCalibrationToolboxforMatlab."ComputationalVisionattheCaliforniaInstituteof
Technology.CameraCalibrationToolboxforMATLAB.
[4]Bradski,G.,andA.Kaehler.LearningOpenCV:ComputerVisionwiththeOpenCVLibrary.Sebastopol,CA:
O'Reilly,2008.

SeeAlso
CameraCalibrator|SingleCameraCalibrationApp

RelatedExamples

EvaluatingtheAccuracyofSingleCameraCalibration

http://www.mathworks.com/help/vision/ug/cameracalibration.html?requestedDomain=www.mathworks.com

4/5

8/11/2016

WhatIsCameraCalibration?MATLAB&Simulink

MeasuringPlanarObjectswithaCalibratedCamera

StructureFromMotionFromTwoViews

http://www.mathworks.com/help/vision/ug/cameracalibration.html?requestedDomain=www.mathworks.com

5/5

You might also like