Physics-Based Vision

1
??????????????_at_?????

• ???
• ?? ???? ?
• ????(??)?????????????
• ????
• ??????????????
• ?????????? --?????????--
• ??????????????1998

2
Physics-Based Vision

• ? ??????,?????????????????????,???????????
• ??????????????(??????????????????)???????????

Observer (Camera)
???
??
??????? ??????????
??
3
Color Image Formation
Observer (Camera)
???
????
??????? ??????????
??
?????????????????????3? ???????????????????
4
Color Image Formation(??)
???
Observer (Camera)
??
??????? ??????????
??
5
????????
Blackbody Radiator?????
Fluorescent light???
??????????????? (Correlated Color Temperature
CCT)??????? ???????????(Kelvin K).
6
????????
Blackbody Radiator
CCT 6200K
CCT 2600K
Fluorescent light
???????????????? ?????????????? K????(????273
)
7
Color Image Formation(??)
??
???
Observer (Camera)
??????? ??????????
??
????????????????
8
Reflectance of Gray
???????????????????????
9
Reflectance of Human Skin
??
??
???????
???
??
??
10
Reflectance of Vegetation Soil
??
?
???????????????????????
11
Color Image Formation(???)
???
Observer (Camera)
??
??????? ??????????
??
12
Observer/Sensor
Camera Response
Eye Response
? Reflected light spectrum is represented by a 3
element vector
???????????????????? ??????????RGB3???????????
13
???(Color System)

• ??????????????????????????????????????????????
  (CIE Color Standard 1931)
• CIE standardized (Commission Internationale de
  LEclairage) ? ???????

14
CIE???

• CIE?imaginary????(X, Y, Z)??

15
Measured vs. CIE???

• measured basis
• ???
• ?????
• ???????

• CIE transformed basis
• ?????????
• ????????
• Y????

16
RGB?CIE???
CIE???
RGB???(???)
RGB??? in CIE???
17
??????????????RGB
Colour Cube
Chromaticity Plane ????
Same Color, different brightnesses
18
HSV???
Saturation
Value
Hue
19
Color Spaces
Models Models Applications
Colorimetirc XYZ(????) Colorimetric calculations
Device-oriented Non-uniform spacesRGB, YIQ, YCC, ... Storage, processing, analysis, coding, color TV
Device-oriented Uniform spacesL a b, L u v, ... Color difference evaluation, analysis, color management systems
User-oriented HSI, HSV, HSL, I1I2I3, ... Human color perception, computer graphics
Munsell Human visual system
20
RGB to YIQYUV

• YIQ???(???????Used in NTSC National Television
  Systems Committee)
• Y???, I Q? (Ired/green,Qblue/yellow)
• Y 0.299 R 0.587 G 0.114 B
• I 0.596 R - 0.275 G - 0.321 B
• Q 0.212 R - 0.528 G 0.311

• YUV???(???????????1982 standard)
• Y 0.299R 0.587G 0.114B
• U 0.492(B Y)
• V 0.877(R Y)

21
CIE???
3???CIE???????2?????????
y
xy???
x
22
????????CIE??????
23
???(Color Discrimination)

• ??????????????

??????????????????????????????
24
????
????I, II? ??????????? I??3?????? ???????????? II?
?????????

• ?RGB??????????
• I,II?????????????????
• ??????????????????

25
???(Color Discrimination)
???????????????? ???????????????10????? ?CIE?????
???????????????????????????????????? ???????????
?????????????????????
26
?????

• ??????????????????????????????????????????????????
  ????????????????????????????
• ?????????????????????(Uniform Chromaticity Scale
  Diagram, UCS Diagram)???

27
CIE1960UCS???

• ?CIE1960UCS???(xy???)?????????????????
• u 4x / (-2x12y3) v 6y / (-2x12y3)
• ?3???X, Y, Z?????
• u 4X / (X15Y3Z) v 6Y / (X15Y3Z) ???
• ?????????????
• ??????????????
• ??????????????????

28
(No Transcript)
29
?????(D.Farnsworth 1957)
??????
30
??

• ?????????????????????
• ?????
• ??????????????????????????????????????????
• ????????????????????????????????????????????????
  ???

31
?????????
Additive (RGB) ????

• ????????????????????
• ?????3???(R)??(G),?(B)??????

32
(No Transcript)
33
??????

• ???????RGB???(Complementary Color)????(C)?????(M)?
  ?(Y)?3????????
• ?? ?????????????????????????????????????
• ????????????????????????????????

34
Subtractive (CMYK) ????

• ?????????????(C)?????(M)??(Y)
• 3?????????????
• ????????????????
• C??R??M??G??Y??B?????
• ????????????????????????????

35
(No Transcript)
36
Color Image Formation(??)
Observer (Camera)
???
????
??
????????????????????????
37
Dichromatic Reflection Model (2???????)??????
Incident Light
Material surface
Absorption(??) Scattering(??) Colorant(???)
?????????2??????????????????????????
38
Example Reflectance of Skin
From Anderson and Parrish
39
?????RGB????????
?For small a the camera RGB will be in the body
direction
?As a becomes large so the RGB moves towards the
surfface direction
40
2???(Dichromatic Plane)
?????RGB? RGBbodya RGBsurface ??bodyRGB?surfac
eRGB?????
RGBs outside this plausible (striped) region can
only occur if we have negative body or surface
contributions. We cannot since light is a
positive quantity
41
Measurement in RGB Space
??????????????
42
Seeing Color Image
?????????????????????3????????????????????

• ???????(shading)
• ????

43
???????1?????
??
???
????????????????????????????
44
???????2?????
?????
????
n
n
e1
e2
??????????????????????? ??????????????????
45
????????

• ?????
• ??????????????????(??)????
• ???????????
• ???
• ???
• ?????
• ?????
• ???

??????????
46
?????????

• ??????
• ????????????????????????????????

???
?????
?????(????)
?????(????)
???
47
Shape From Shading
48
????3D??????
????
?????3D??
49
Gretag Macbeth ColorChecker
50
???????????
Macbeth color checker
????????
?????
Daylight
Tungsten
Redder
51
??????
Illuminant Color Tungsten
Macbeth color checker???
??????
Color Correction
?????????????? ???????????????????
52
??????
Macbeth color checker???
Illuminant Estimation
?
Color Correction
53
Estimating the Illuminant ColorComputational
Color Constancy
?????????????
54
Example of ApplicationColor Correction
Estimation error 1.6
von Kreis Transformation
Re-rendered image to canonical cond. ???????
Image taken underCCT 4700K
Canonical image at CCT 3200K
55
Color Image Segmentation

• Pixel-based techniques(??)
• Region-based techniques(??)
• Edge-based techniques(???)
• Stochastical Model-based techniques(???)
• Physics-based techniques(??)
• Hybrid techniques(??)
• Example modelling skin and segmenting

56
Color Clustering (or Segmentation) by K-means
Algorithm
57
Color Clustering (Segmentation) by K-means
Form K-means clusters from a set of n-dimensional
vectors 1. Set i (iteration count) to 1 2.
Choose randomly a set of K means m1(1), ,
mK(1). 3. For each vector xi, compute
D(xi,mk(ic)), k1,K and assign xi to the
cluster Cj with nearest mean. 4. Increment i by
1, update the means to get m1(i),,mK(i). 5.
Repeat steps 3 and 4 until Ck(i) Ck(i1) for
all k.
58
?????????(K-means Tracker)
????
????
??????

???????????
5??????
?????
??????????????
???????????????
59
??????
?????????????3???????? 3??????????????????? ????
????????????????