Chapter 6 : Color and Shading

1
Chapter 6 Color and Shading

• Computer Vision Lab.
• 2005. 7. 18
• ???
• ???

2
Contents

• 6.1 Some physics of color
• 6.2 The RGB basis for color
• 6.3 Other color bases

3
Some physics of color (1)

• perception of color
• nanometer 10-9 meter

4
Some physics of color (2)

• Sensing illuminated objects
• The sensation, or perception, of an objects
  color depends upon three general factors
• the spectrum of energy
• the spectrum reflectance of the objects surface
• the spectral sensitivity of the sensor
• White light is composed of
• approximately equal energy
• in all wavelengths of the
• visible spectrum.

5
Some physics of color (3)

• Additional factors
• variation of intensity with distance
• variation of intensity with surface normal
• Sensitivity of receptors

6
The RGB basis for color

• RGB
• (28)3 or 16 million color codes
• additive color system (ex black(0,0,0))
• one way to normalize image data
• intensity I (RGB)/3
• normalized red r R/(RGB)
• normalized green g G/(RGB)
• normalized blue b B/(RGB)
• r g b 1

7
Other color bases (1)

• The CMY subtractive color system
• printing on white paper
• abbreviation of Cyan-Magenta-Yellow
• subtractive system for the encoding for
  absorption
• white(0, 0, 0)
• black(255, 255, 255)
• yellow(0 0 255)

8
Other color bases (1)

• HSI Hue-Saturation-Intensity
• HSV Hue-Saturation-Value

9
Other color bases (1)
10
Other color bases (2)

• YIQ and YUV for TV signals
• YIQ encoding
• luminance Y 0.3R 0.59G 0.11B
• R-cyan I 0.6R 0.28G 0.32B
• magenta-green Q 0.21R - 0.52G - 0.31B
• YUV encoding
• Digital video products and compression algorithms
  such as JPEG and MPEG
• Y 0.3R 0.59G 0.11B
• U 0.493(B - Y)
• V 0.877(R Y)

11
Other color bases (3)

• Using color for classification
• finding a human face
• prone to error
• interpretation of an individual pixel
• specular reflection
• particular regions of the color space

12
Contents

• 6.4 Color histograms
• 6.5 Color Segmentation

13
Color Histograms

• Why do make the histograms
• Histogram is fast and easy to compute.
• Size can easily be normalized so that different
  image histograms can be compared.
• Can match color histograms for database query or
  classification.

14
Color Histograms

• Creating a histogram
• Gray

15
Color Histograms

• How to make a color histograms
• Make 3 histograms and concatenate them
• using higher n bits of each RGB
• ex) if n 2 then 26 64 bins
• Can normalize histogram to hold frequencies so
  that bins total 1.0

16
Color Histograms

• How to make a color histograms
• Make 3 histograms

17
Color Histograms

• How to make a color histograms
• using higher n bits of each RGB

4 bit 48 bins
18
Color Histograms

• Histogram Matching for Color Object Recognition
• Swain and Ballards

H(I)
H(M)
19
Color Segmentation

• Segmentation is the process of identifying based
  on common properties
• These properties could include intensity, color,
  texture, etc.
• Thresholding grayscale images can be useful for
  segmentation

20
Color Segmentation

• Application

21
Color Segmentation

• Application
• Training samples of skin color in RGB-space
• RGB-space divided into skin versus non-skin
• Input image pixels labeled skin versus non-skin
• Connected components extracted
• Small regions thrown away, largest one kept
• Some regions merged into largest face region

22
Color Segmentation

• Application

(left) input video frame, (center) pixels
classified according to RGB space, (right)
largest connected component with aspect similar
to a face (all work contributed by Vera Bakic)
23
Color Segmentation

• (a) Original image.
• (b) Pixels detected as skin.
• (c) Segmentation after histogram clustering.
• (d) Label erosion of size 2 of c.
• (e) Label dilation of size 2 of d.
• (f) Watershed segmentation using the markers of e.

24
Shading

• Several factors affect how am image is viewed or
  captures
• specularity of the surface
• distance of the light source and the camera
• angle of the light source on the object surface
• all play a role in the perception of an object

25
Shading

• Radiation from one light source
• Received i n ? s.
• Cj/Aj cos?j
• ?j angle between nj and s

26
Shading

• Diffuse reflection
• Diffuse reflected i nj ? s
• A diffuse reflecting surface reflects light
  uniformly in all directions
• Uniform brightness for all viewpoints of a planar
  surface.

27
Shading

• Diffuse reflection from lambertain objects

28
Shading

• Sepecular reflection

R is the ray of reflection V is direction
from the surface toward the viewpoint ? is
the shininess parameter
29
Shading

• Darkening with distance
• The intensity of light received by any object
  surface will decrease with square of its distance
  from the source.