SIMULTANEOUS GEOMETRIC AND COLORIMETRIC CAMERA CALIBRATION Ilmenau, - - PowerPoint PPT Presentation

▶
simultaneous geometric and colorimetric camera calibration
SMART_READER_LITE
LIVE PREVIEW

SIMULTANEOUS GEOMETRIC AND COLORIMETRIC CAMERA CALIBRATION Ilmenau, - - PowerPoint PPT Presentation

Oct 23, 2022 35 likes •188 views

SIMULTANEOUS GEOMETRIC AND COLORIMETRIC CAMERA CALIBRATION Ilmenau, 7th October 2010 Daniel Kapusi, Philipp Prinke, Rainer Jahn, Darko Vehar, Rico Nestler, Karl-Heinz Franke Zentrum fr Bild- und Signalverarbeitung e. V. www.zbs-ilmenau.de

slide-1
SLIDE 1

SIMULTANEOUS GEOMETRIC AND COLORIMETRIC CAMERA CALIBRATION

Ilmenau, 7th October 2010

Daniel Kapusi, Philipp Prinke, Rainer Jahn, Darko Vehar, Rico Nestler, Karl-Heinz Franke

Zentrum für Bild- und Signalverarbeitung e. V.

www.zbs-ilmenau.de

slide-2
SLIDE 2

Outline

SIMULTANEOUS GEOMETRIC AND COLORIMETRIC CAMERA CALIBRATION

  • 07. 10. 2010

2

  • 16. Farbworkshop
  • Motivation
  • Calibration Target
  • Target Feature Detection
  • Geometric Calibration
  • Stereo Rectification
  • Color Calibration
  • Conclusion
slide-3
SLIDE 3

Conclusion Color Calibration Stereo Rectification Geometric Calibration Target Feature Detection Calibration Target Motivation

Application Scenario: Robot Work Area

D D D D D D D D

stereo vision camera robot

D D D D

monocular vision camera human

Motivation

3

Vision based safety system: Avoiding collisions between robots and humans! 3D scene modeling

  • Redundant scene observation

from multiple points of view

  • Stereo vision for more robust
  • bject recognition

Geometric and Colorimetric Camera Calibration

Requirements

  • Unified coordinate system for

all sensors

  • Unified colorimetric perception

between all sensors

  • 07. 10. 2010
  • 16. Farbworkshop
slide-4
SLIDE 4

Conclusion Color Calibration Stereo Rectification Geometric Calibration Target Feature Detection Calibration Target Motivation

Chessboard Calibration Target With Colorimetric Features

  • 07. 10. 2010
  • 16. Farbworkshop

4

Calibration Target

Geometric features

  • Different parity in number of

edges in horizontal and vertical direction [1] Orientation is clearly detectable Colorimetric features

  • 24 circles filled with well distributed

reference colors (natural and primary colors as well as grey- balanced fields) in the style of the Macbeth [2] color checker

[2] C. S. McCamy, H. Marcus, and J. G. Davidson, "A Color-Rendition Chart,“ Journal of Applied Photographic Engineering, vol. 2, no. 3, pp. 95-99, Summer 1976. [1] Jean-Yves Bouguet. (2010, July) Camera Calibration Toolbox for Matlab. [Online]. http://www.vision.caltech.edu/bouguetj/calib_doc/index.html

slide-5
SLIDE 5

Conclusion Color Calibration Stereo Rectification Geometric Calibration Target Feature Detection Calibration Target Motivation

Values Of The ColorChecker

  • 07. 10. 2010
  • 16. Farbworkshop

5

Calibration Target

[3] Danny Pascale, "RBG coordinates of the Macbeth ColorChecker," The BabelColor Company, Montreal, Quebec, Canada, Comparison 2006.

slide-6
SLIDE 6

Conclusion Color Calibration Stereo Rectification Geometric Calibration Target Feature Detection Calibration Target Motivation

Detecting Edges On The Chessboard Using OpenCV

  • 07. 10. 2010
  • 16. Farbworkshop

6

Target Feature Detection

  • 1. Step: rough edge detection [4]
  • Adaptive threshold binarization
  • Separate black squares by dilatation of white squares
  • Find and simplify contours to 4 edge-points at each black square
  • 2. Step: sub pixel accurate edge detection [4]
  • Using the fact, that the dot product of two orthogonal vectors is zero
  • Consider all points within a neighborhood around the real corner position Q
  • Iterative solution of a linear system of equations

) ( ) (     q p P I

i T i

pi – position vector to point Pi q – position vector to point Q

[4] Gary Bradski and Adrian Kaehler, Learning OpenCV. Sebastopol, USA/CA: O´Reilly Media, 2008

slide-7
SLIDE 7

Conclusion Color Calibration Stereo Rectification Geometric Calibration Target Feature Detection Calibration Target Motivation

Detecting Edges On The Chessboard Using OpenCV

  • 07. 10. 2010
  • 16. Farbworkshop

7

Target Feature Detection Multipose-Calibration

slide-8
SLIDE 8

Conclusion Color Calibration Stereo Rectification Geometric Calibration Target Feature Detection Calibration Target Motivation

Detecting Color Features

  • 07. 10. 2010
  • 16. Farbworkshop

8

Target Feature Detection

Calculating the start color value from each square

  • Middle position of the chessboard square
  • Median value from direct neighborhood

Recursive region growing

  • Using a uniform criterion as a tolerance threshold

Average the color values from the segmented region

slide-9
SLIDE 9

Conclusion Color Calibration Stereo Rectification Geometric Calibration Target Feature Detection Calibration Target Motivation Image World Camera

Camera Parameters

  • 07. 10. 2010
  • 16. Farbworkshop

9

Intrinsic camera parameters Hx, Hy – Principal point coordinates F – Focal length Extrinsic camera parameters R – Rotation matrix t – Translation vector

[5] Roger Y. Tsai, "An Efficient and Accurate Camera Calibration Technique for 3D Machine Vision," IEEE Journal of Robotics and Automation, vol. 3, no. 4, pp. 323-344, August 1987. [6] Zhengyou Zhang, "A Flexible New Technique for Camera Calibration," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 22, no. 11, pp. 1330-1334, November 2000.

Coordinate transformation World Camera Image

  P

t R M p   

  P

t R 

           1

y y x x

H f H f M Geometric Calibration

slide-10
SLIDE 10

Conclusion Color Calibration Stereo Rectification Geometric Calibration Target Feature Detection Calibration Target Motivation

Brown-Conrady-Distortion-Model

  • 07. 10. 2010
  • 16. Farbworkshop

10 radial distortion tangential distortion

  • ptical centre

Distortion parameters k1, k2, k3 - radial distortion coefficients p1, p2

  • tangential distortion coefficients

tangential

 

) 2 ( 2 1

2 2 2 1 6 3 4 2 2 1 u u u u v

x r p y x p r k r k r k x x        

 

u u u u v

y x p y r p r k r k r k y y

2 2 2 1 6 3 4 2 2 1

2 ) 2 ( 1        

2 2 2 u u

y x r  

radial radial distortion + tangential distortion

[7] Alexander Eugen Conrady, "Decentering lens systems," Monthly notices of the Royal Astronomical Society, vol. 79, pp. 384-390, April 1919. [8] Duane C. Brown, "Decentering Distortion of Lenses," Photometric Engineering, vol. 32, no. 3, pp. 444-462, 1966.

Geometric Calibration

slide-11
SLIDE 11

Conclusion Color Calibration Stereo Rectification Geometric Calibration Target Feature Detection Calibration Target Motivation

D D

S t e r e

  • V

i s i

  • n

B V

  • S

y s t e m

D

MonoVision BV-System

D

MonoVision BV-System

D

MonoVision BV-System

D

MonoVision BV-System

D D

StereoVision BV-System

D D

StereoVision BV-System

D D

StereoVision BV-System

Overall Calibration Procedure

  • 07. 10. 2010
  • 16. Farbworkshop

11 D D

StereoVision BV-System

D

MonoVision BV-System

D D

StereoVision BV-System

D

MonoVision BV-System

D D

StereoVision BV-System

D

MonoVision BV-System

D D

StereoVision BV-System

R+T R+T R+T R+T R+T R+T R+T R+T R+T R+T R+T R+T R+T R+T R+T

Geometric Calibration

D

MonoVision BV-System

slide-12
SLIDE 12

Conclusion Color Calibration Stereo Rectification Geometric Calibration Target Feature Detection Calibration Target Motivation

Stereo Rectification[4]

  • 07. 10. 2010
  • 16. Farbworkshop

12

Stereo Rectification

Extraction of depth information

  • Assuming a standard stereo geometry
  • Determination of the horizontal mismatch (disparity)

from a stereoscopic view (Correspondence analysis)

  • Calculating the distance by simple triangulation

[4] Gary Bradski and Adrian Kaehler, Learning OpenCV. Sebastopol, USA/CA: O´Reilly Media, 2008

slide-13
SLIDE 13

Conclusion Color Calibration Stereo Rectification Geometric Calibration Target Feature Detection Calibration Target Motivation

  • 07. 10. 2010
  • 16. Farbworkshop

13

Stereo Rectification[4]

Stereo Rectification left right unrectified image rectified images

[4] Gary Bradski and Adrian Kaehler, Learning OpenCV. Sebastopol, USA/CA: O´Reilly Media, 2008

slide-14
SLIDE 14

Conclusion Color Calibration Stereo Rectification Geometric Calibration Target Feature Detection Calibration Target Motivation

Color Calibration

  • 07. 10. 2010
  • 16. Farbworkshop

14

Color Calibration

RGB act

C

actual color valence

RGB nom

C

Calculation of Transformation Rule nominal color valence

Target based global color calibration[9]

  • Corresponding nominal and actual color values
  • Calculation of the transformation rule which causes the smallest middle aberration

[9] ZBS e. V. (2007, April) zbs CCal-Bibliothek Benutzerdokumentation. [Online]. http://www.zbs-ilmenau.de/intern/ccalc/ZBSColCalib.pdf

Linear transformation models

  • Scaling (+Offset)
  • Linear Transformation
  • Linear Regression

Non-linear transformation model

  • Polynomial regression
slide-15
SLIDE 15

Conclusion Color Calibration Stereo Rectification Geometric Calibration Target Feature Detection Calibration Target Motivation

Conclusion

  • 07. 10. 2010
  • 16. Farbworkshop

15

Conclusion

D D D D D D D D

stereo vision camera robot

D D D D

monocular vision camera human

  • Using a chessboard target with integrated color markers
  • Sub pixel accurate edge detection of the chessboard squares
  • Calculating intrinsic camera parameters, distortion coefficients

and geometric relations to adjacent sensors from multiple target poses

  • Calculate rectification parameters for stereo vision systems
  • Target based color calibration

Available as an ANSI C conform software library Geometric and Colorimetric Camera Calibration Requirements to 3D scene modeling

  • Unified coordinate system for all sensors
  • Unified colorimetric perception between all sensors