Objectives_template

  Module 6: Pinhole camera model

  Lecture 30: Intrinsic camera parameters, Perspective projection using homogeneous coordinates
 
 
 The Lecture Contains:
 
  Pinhole camera model
 
  6.1 Intrinsic camera parameters
 
A. Perspective projection using homogeneous coordinates
 
  B. Principal-point offset
 
  C. Image-sensor characteristics
 
 
 

file:///D|/...0(Ganesh%20Rana)/MY%20COURSE_Ganesh%20Rana/Prof.%20Sumana%20Gupta/FINAL%20DVSP/lecture%2030/30_1.htm[12/31/2015 1:23:40 PM]

Objectives_template

  Module 6: Pinhole camera model

  Lecture 30: Intrinsic camera parameters, Perspective projection using homogeneous coordinates
  
  Pinhole camera model
  In this section, we describe the image acquisition process known as the pinhole camera model ,
 
which is widely used. More specifically, we first discuss the model that integrates the internal or
 
  intrinsic camera parameters, such as the focal length and the lens distortion. Secondly, we extend
  the presented simple camera model to integrate external or extrinsic camera parameters
  corresponding to the position and orientation of the camera.
  
  6.1 Intrinsic camera parameters
  The pinhole camera model defines the geometric relationship between a 3D point and its 2D
  corresponding projection onto the image plane. When using a pinhole camera model, this geometric
 
mapping from 3D to 2D is called a perspective projection. We denote the center of the perspective
projection (the point in which all the rays intersect) as the optical center or camera center and the
line perpendicular to the image plane passing through the optical center as the optical axis (see
Figure  6.2 ). Additionally, the intersection point of the image plane with the optical axis is called the
 
principal point. The pinhole camera that models a perspective projection of 3D points onto the image
plane can be described as follows.
 
 

file:///D|/...0(Ganesh%20Rana)/MY%20COURSE_Ganesh%20Rana/Prof.%20Sumana%20Gupta/FINAL%20DVSP/lecture%2030/30_2.htm[12/31/2015 1:23:40 PM]

Objectives_template

  Module 6: Pinhole camera model

  Lecture 30: Intrinsic camera parameters, Perspective projection using homogeneous coordinates
  
  A. Perspective projection using homogeneous coordinates
 
Let us consider a camera with the optical axis being collinear to the - axis and the optical
 
  center being located at the origin of a 3D coordinate system (see Figure 6.1).
 
 
 
 
 
 
 
 

Figure 6.1  The ideal pinhole camera model describes the relationship between a 3D point
and its corresponding 2D projec tion .

 
 

file:///D|/...0(Ganesh%20Rana)/MY%20COURSE_Ganesh%20Rana/Prof.%20Sumana%20Gupta/FINAL%20DVSP/lecture%2030/30_3.htm[12/31/2015 1:23:40 PM]

Objectives_template

  Module 6: Pinhole camera model

  Lecture 30: Intrinsic camera parameters, Perspective projection using homogeneous coordinates
  
  The projection of a 3D world point onto the image plane at pixel position can be
  written as
 
  (6.3)
 
  where f denotes the focal length. To avoid such a non-linear division operation, the previous relation
  can be reformulated using the projective geometry framework, as
  (6.4)
 
This relation can be the expressed in matrix notation by  
 
 
  (6.5)

  where is the homogeneous scaling factor.  

   
 

file:///D|/...0(Ganesh%20Rana)/MY%20COURSE_Ganesh%20Rana/Prof.%20Sumana%20Gupta/FINAL%20DVSP/lecture%2030/30_4.htm[12/31/2015 1:23:41 PM]

Objectives_template

  Module 6: Pinhole camera model

  Lecture 30: Intrinsic camera parameters, Perspective projection using homogeneous coordinates
  
  B. Principal-point offset
 
  
  Most of the current imaging systems define the origin of the pixel coordinate system at the top-left
  pixel of the image. However, it was previously assumed that the origin of the pixel coordinate system
  corresponds to the principal point , located at the center of the image (see Figure 6.2(a)).
  A conversion of coordinate systems is thus necessary. Using homogeneous coordinates, the
  principal-point position can be readily integrated into the projection matrix. The perspective projection
  equation becomes now
 
 
(6.6)
 

   
C. Image-sensor characteristics  
 
To derive the relation described by Equation ( 6.6 ), it was implicitly assumed that the pixels of the
  image sensor are square, i.e., aspect ratio is 1 : 1 and pixels are not skewed. However, both
assumptions may not always be valid. First, for example, an NTSC TV system defines non-square
pixels with an aspect ratio of 10 : 11. In practice, the pixel aspect ratio is often provided by the
image-sensor manufacturer. Second, pixels can potentially be skewed, especially in the case that
the image is acquired by a frame grabber.
 
 

file:///D|/...0(Ganesh%20Rana)/MY%20COURSE_Ganesh%20Rana/Prof.%20Sumana%20Gupta/FINAL%20DVSP/lecture%2030/30_5.htm[12/31/2015 1:23:41 PM]

Objectives_template

  Module 6: Pinhole camera model

  Lecture 30: Intrinsic camera parameters, Perspective projection using homogeneous coordinates
  
  In this particular case, the pixel grid may be skewed due to an inaccurate synchronization of the pixel-
  sampling process. Both previously mentioned imperfections of the imaging system can be taken into
 
account in the camera model, using the parameters ? and t, which model the pixel aspect ratio and
 
skew of the pixels, respectively (see Figure  6.2(b) ). The projection mapping can be now updated as
 
 
  (6.7)
 
 
  with being a 3D point defined with homogeneous coordinates. In practice, when
  employing recent digital cameras, it can be safely assumed that pixels are square and non-
  skewed . The projection matrix that incorporates the intrinsic parameters is denoted as K.
The all zero element vector is denoted by .
 

Figure 6.2  (a) The image and camera coordinate system. (b) Non-ideal image sensor
with non-square, skewed pixels.
 
 

file:///D|/...0(Ganesh%20Rana)/MY%20COURSE_Ganesh%20Rana/Prof.%20Sumana%20Gupta/FINAL%20DVSP/lecture%2030/30_6.htm[12/31/2015 1:23:41 PM]