Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

computer_vision_teorico > Camera Calibration > Flashcards

Camera Calibration Flashcards

1
Q

Camera Parameters and Calibration:
What is Luminance (cd/m²)?

A
  • measurement of the photometric luminous intensity per unit area of light
  • its the amount of light that goes through a particular area, and falls within a given solid angle
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What is Chrominance?

A
  • the way a certain amout of light is distributed among the visible spectrum
  • greyscale image has no color information
  • any RGB triplet has no chrominance information
  • chrominance has no luminance information but is used together with it to describe colored image
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

How is RGB & YUV defined?

A

-Y is luminance and is calculated from 3 coefficients and the R, G and B values
-U and V are the chrominance

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What is Image Processing Pipeline:

A

-This processing can be performed on the YUV or RGB
components depending on the system.
-Depending on the system, more or less image parameters may be available for the user to control. Also, some of these parameters (namely brightness, contrast and saturation) are also intrinsic original image characteristics apart from being
externally controllable parameters.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

How many Image parameters are there?

A

6

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Numerate the Image parameters

A
  • Brightness:
  • Contrast:
  • White Balance:
  • Saturation:
  • Gamma:
  • Sharpness:
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What defines Brightness? (as an intrinsic original image characteristic)

A
  • average amount of light that is integrated over the image during exposure time
  • is a constant (or offset) that can be added (subtracted) from the luminance component of the image (as a controllable parameter)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What is Contrast? (as an intrinsic image characteristic)

A
  • difference in luminance (or color) along the 2D space - most used definition
  • difference in the color and brightness of the object in the same field of view - real world
  • the faster and higher the luminance changes along the space the higher the contrast is
  • gain control function of the luminance of the image
  • normally combined with Brightness in a single transfer function
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

How is Contrast calculated?

A

luminance difference / average luminance

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What is white Balance? (as a controllable parameter)

A
  • global adjustment of the intensities of the colors (red, green and blue primary colors)
  • goal is to render specific colors correctly- particularly neutral colors
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What is Saturation? (as an intrinsic image characteristic)

A
  • distribution of light accross the spectrum of different wavelengths
  • amount of white you have blended into a pure color
    -If the light intensity declines, then, as a result, the saturation also decline.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

How can the purest (most saturated) color be obtained?

A

It is obtained when using a single wavelength at a high intensity(laser light is a good example)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

How do you remove the saturation of an image? (as a controllable parameter)

A

By adding white to the original colors (this is the same as changing the gain of the U and V chromatic components).

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What is Gamma?

A
  • nonlinear operation used to code and decode luminance values
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

How is Gamma calculated ?

A

Vout = A*Vin^δ
where A is a constant and the input and output values are non-negative real values.
In most cases A = 1, and inputs and outputs are typically in the range 0–1.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What is Sharpness? (as a controllable parameter)

A

-The measure of the energy frequency spatial distribution over the image
-Not all devices provide access to this parameter.
- basically allows the control of the cut-off frequency of a low pass
spatial filter.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

In which cases is sharpness very useful?

A

if the image is afterward intended to be decimated, since it allows to prevent spatial aliases artifacts.

18
Q

Is Exposure time a controllable parameter of the image device?

A

It may or may not be

19
Q

What happens if the brightness is too high?

A

Overexposure may occur which will white saturate part
or the totality of the image.

20
Q

What is the maximum possible contrast of an image denominated?

A

Contrast ratio or dynamic range.

21
Q

Is it common that contrast and brightness are actually a combined single transfer
function?

A

Yes

22
Q

What kind of Camera parameters are there?

A

Extrinsic:
- define the location and orientation of the camera reference with respect to a known reference
Intrinsic:
- link the image pixel coordinates with the corresponding coordinates in the camera reference frame

23
Q

How do you accommodate the camera internal
parameters and geometry and its position and
posture in the real worl?

A

Using geometrical transformations.
Object coordinates (3D) -> World Coordinates (3D) -> Camera Coordinates (3D) -> Image Plane Coordinates (2D) -> Pixel Coordinates (2D, int)

24
Q

What is the pinhole model?

A

In this model, the lens is replaced by a very narrow opening
(pinhole) through which lights go through directly into the image acquisition plane.
The point that is stroked by a light ray going through the pin hole in the direction of the lens main axis is called the image plane origin.

25
Q

In most digital image processing tasks from which model can the optical system be approximated by?

A

By the pinhole model.

26
Q

In the pinhole model where does the acquisition plane lies?

A

At the focal distance from the pinhole.

27
Q

Are the image optical centre and sensor centre normally aligned?

A

No

28
Q

How are the x and y from the screen obtained?

A

xscreen = fx(X/Z) + cx
yscreen = fy(Y/Z) + cy
where fx and fy are derived from focal distance but take
in consideration pixel size of the sensor

29
Q

What are Homogenous Coordinates?

A

-It allows any point in a 3D space to be represented by
a multitude of equivalent matrixes.
-It also allows us to represent any point in a plane
which is at an infinite distance from the origin. Such
representation will have its w = 0

30
Q

Which parameters can rotation of a vector defined by two points in a Cartesian system be obtained from?

A

an : coordinates of the point to be projected;
cn : the pinhole coordinates;
Dn : resulting rotated vector.

31
Q

Optic basics:

A
  • Optical lenses
    • Formation of image: 1/S1 + 1/S2 = 1/f
    • Spherical Aberration
      • results in multiple focus points
      • can be corrected using the simplified Browns distortion model
32
Q

What do most simple image acquisition systems use?

A

A single converging lens as its optical interface

33
Q

What is shperical aberration?

A

Image deformation resulting from the fact that most
lenses have a spherical surface cut

34
Q

What types of spherical aberration are there?

A

-a barrel form (left side) -most common one and normally increases with diminishing focal distance.
-pincushion effect (center)
-Mustache distortion (right).

35
Q

How many parameters does the OpenCV camera model have?

A

15 :
- 4 intrinsic parameters:
- focal distance: fx, fy
- optical center: cx, cy
- 6 extrinsic parameters:
- rotation: rx, ry, rz
- translation: tx, ty, tz
- 5 distortion parameters:
- lens distortion:
- radial distortion: k1, k2, k3
- tangential distortion: p1, p2

36
Q

Are there other models?

A

Yes, for example: Tsai, Heikkila, Zhang

37
Q

What is the camera calibration?

A

Given 3D-2D correspondences is it possible to
define a system of equation that might be solved
numerically.

38
Q

What is the single camera calibration problem?

A

Occurrence of Outliers - Only some matches are mutually consistent

39
Q

How do you remove the outliars?

A

RANSAC - “RANdom SAmple Consensus” might
be used to remove outliers

40
Q

What is RANSAC?

A
  • example of a voting based fitting scheme
  • each hypothesis gets voted on by each data point, best hypothesis wins
41
Q

What are some Camera calibration functions using OpenCV?

A

-cvFindChessboardCorners: internal corners of
the chessboard
– Optional: cvFindCornerSubPix: Refines the
corner locations
* cvCalibrateCamera2: find intrinsic and extrinsic
parameters from several views of chessboard.
* solvePnP: finds an object pose from 3D-2D point
correspondences.

computer_vision_teorico (12 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions