Vladimir Botchko [email protected]
description
Transcript of Vladimir Botchko [email protected]
![Page 1: Vladimir Botchko botchko@lut.fi](https://reader036.fdocuments.net/reader036/viewer/2022070502/56814c29550346895db92e87/html5/thumbnails/1.jpg)
1
Vladimir Botchko [email protected]
Lecture 5. Color Image ProcessingLecture 5. Color Image Processing
Lappeenranta University of Technology (Finland)
![Page 2: Vladimir Botchko botchko@lut.fi](https://reader036.fdocuments.net/reader036/viewer/2022070502/56814c29550346895db92e87/html5/thumbnails/2.jpg)
2
Color Image ProcessingColor Image Processing
Fundamentals Color models Pseudocolor image processing Full color image processing Color transformations Smoothing and Sharpening Color segmentation Noise in color images Color image compression Multispectral image processing
![Page 3: Vladimir Botchko botchko@lut.fi](https://reader036.fdocuments.net/reader036/viewer/2022070502/56814c29550346895db92e87/html5/thumbnails/3.jpg)
3
Fundamentals
Colors in the visible range of wavelengths (upper left), mixtures of light (additive primaries) (upper right) and color bars used in analysis.
![Page 4: Vladimir Botchko botchko@lut.fi](https://reader036.fdocuments.net/reader036/viewer/2022070502/56814c29550346895db92e87/html5/thumbnails/4.jpg)
4
Color models
Relative color gamuts of a dipslay and a printer in XYZ chromaticity coordinate system (right).
Left - XYZ color space.
![Page 5: Vladimir Botchko botchko@lut.fi](https://reader036.fdocuments.net/reader036/viewer/2022070502/56814c29550346895db92e87/html5/thumbnails/5.jpg)
5
Color Image ProcessingColor Image Processing
Fundamentals Color models Pseudocolor image processing Full color image processing Color transformations Smoothing and Sharpening Color segmentation Noise in color images Color image compression Multispectral image processing
![Page 6: Vladimir Botchko botchko@lut.fi](https://reader036.fdocuments.net/reader036/viewer/2022070502/56814c29550346895db92e87/html5/thumbnails/6.jpg)
6
Color models
http://cvision.ucsd.edu/index.htm
![Page 7: Vladimir Botchko botchko@lut.fi](https://reader036.fdocuments.net/reader036/viewer/2022070502/56814c29550346895db92e87/html5/thumbnails/7.jpg)
7
Color models
RGB system, HSI (or HSV) system (right) (I-intensity, V value)
![Page 8: Vladimir Botchko botchko@lut.fi](https://reader036.fdocuments.net/reader036/viewer/2022070502/56814c29550346895db92e87/html5/thumbnails/8.jpg)
8
Color models
Three match curves. RGB system (CIE 1931)(left), XYZ system (CIE 1931)(right)
![Page 9: Vladimir Botchko botchko@lut.fi](https://reader036.fdocuments.net/reader036/viewer/2022070502/56814c29550346895db92e87/html5/thumbnails/9.jpg)
9
Color models
RGB space. The right image is a rotated left image (for correspondence: BL is black, W is white).
![Page 10: Vladimir Botchko botchko@lut.fi](https://reader036.fdocuments.net/reader036/viewer/2022070502/56814c29550346895db92e87/html5/thumbnails/10.jpg)
10
Hue, saturation, intensity system
Y aR cG dBr g b
r g bW PW S
r g b
UV|||W|||
arccos[( / ) ( / ) ( / ) ]
min( , , )
/
26 1 3 1 3 1 3
1 3
2 2 2 1 2
![Page 11: Vladimir Botchko botchko@lut.fi](https://reader036.fdocuments.net/reader036/viewer/2022070502/56814c29550346895db92e87/html5/thumbnails/11.jpg)
11
Color models
Chromaticity
r R R G Bg G R G Bb B R G B
UV|W|/ ( )/ ( )/ ( )
g
r0r g b 1b r b 1
w( / , / )1 31 3
![Page 12: Vladimir Botchko botchko@lut.fi](https://reader036.fdocuments.net/reader036/viewer/2022070502/56814c29550346895db92e87/html5/thumbnails/12.jpg)
12
Color models
Multitriangle representation (left) Luminance, chromaticity (right)
![Page 13: Vladimir Botchko botchko@lut.fi](https://reader036.fdocuments.net/reader036/viewer/2022070502/56814c29550346895db92e87/html5/thumbnails/13.jpg)
13
Color models
Karhunen-Loev system
![Page 14: Vladimir Botchko botchko@lut.fi](https://reader036.fdocuments.net/reader036/viewer/2022070502/56814c29550346895db92e87/html5/thumbnails/14.jpg)
14
Color Image ProcessingColor Image Processing
Fundamentals Color models Pseudocolor image processing Full color image processing Color transformations Smoothing and Sharpening Color segmentation Noise in color images Color image compression Multispectral image processing
![Page 15: Vladimir Botchko botchko@lut.fi](https://reader036.fdocuments.net/reader036/viewer/2022070502/56814c29550346895db92e87/html5/thumbnails/15.jpg)
15
Pseudocoloring
Myocardial perfusion study. Left is a heart attack (blue region increased), right is normal.
![Page 16: Vladimir Botchko botchko@lut.fi](https://reader036.fdocuments.net/reader036/viewer/2022070502/56814c29550346895db92e87/html5/thumbnails/16.jpg)
16
Pseudocoloring. X-rays.
a
![Page 17: Vladimir Botchko botchko@lut.fi](https://reader036.fdocuments.net/reader036/viewer/2022070502/56814c29550346895db92e87/html5/thumbnails/17.jpg)
17
Pseudocoloring Right – three images: elevation relief (upper left), the color coded
magnetic field (higher values are yellowish) (upper right), the composition of first two. Left – underpainting revealed through color dipslay (Prof. L. MacDonald, Derby University,GB).
![Page 18: Vladimir Botchko botchko@lut.fi](https://reader036.fdocuments.net/reader036/viewer/2022070502/56814c29550346895db92e87/html5/thumbnails/18.jpg)
18
Thematic classification of six-band satellite imagery using a minimum distance classifier
![Page 19: Vladimir Botchko botchko@lut.fi](https://reader036.fdocuments.net/reader036/viewer/2022070502/56814c29550346895db92e87/html5/thumbnails/19.jpg)
19
Color Image ProcessingColor Image Processing
Fundamentals Color models Pseudocolor image processing Full color image processing Color transformations Smoothing and Sharpening Color segmentation Noise in color images Color image compression Multispectral image processing
![Page 20: Vladimir Botchko botchko@lut.fi](https://reader036.fdocuments.net/reader036/viewer/2022070502/56814c29550346895db92e87/html5/thumbnails/20.jpg)
20
Painting Restoration. A Queen house , London. The part of painting was copied from another painting (upper right) and
used for restoration of the lost painting part.
![Page 21: Vladimir Botchko botchko@lut.fi](https://reader036.fdocuments.net/reader036/viewer/2022070502/56814c29550346895db92e87/html5/thumbnails/21.jpg)
21
Color Image ProcessingColor Image Processing
Fundamentals Color models Pseudocolor image processing Full color image processing Color transformations Smoothing and Sharpening Color segmentation Noise in color images Color image compression Multispectral image processing
![Page 22: Vladimir Botchko botchko@lut.fi](https://reader036.fdocuments.net/reader036/viewer/2022070502/56814c29550346895db92e87/html5/thumbnails/22.jpg)
22
Color segmentation
Image segmentation based on color feature: burnt forest area, forest fire, dead forest (brown).
![Page 23: Vladimir Botchko botchko@lut.fi](https://reader036.fdocuments.net/reader036/viewer/2022070502/56814c29550346895db92e87/html5/thumbnails/23.jpg)
23
Color Image ProcessingColor Image Processing
Fundamentals Color models Pseudocolor image processing Full color image processing Color transformations Smoothing and Sharpening Color segmentation Noise in color images Color image compression Multispectral image processing
![Page 24: Vladimir Botchko botchko@lut.fi](https://reader036.fdocuments.net/reader036/viewer/2022070502/56814c29550346895db92e87/html5/thumbnails/24.jpg)
24
Color image compression.
Original color image (upper left), compressed image (upper right), error histogram in compression (the error is a delta E – the smallest color noticible difference) and error image (large error values are white).
![Page 25: Vladimir Botchko botchko@lut.fi](https://reader036.fdocuments.net/reader036/viewer/2022070502/56814c29550346895db92e87/html5/thumbnails/25.jpg)
25
Color Image ProcessingColor Image Processing
Fundamentals Color models Pseudocolor image processing Full color image processing Color transformations Smoothing and Sharpening Color segmentation Noise in color images Color image compression Multispectral image processing
![Page 26: Vladimir Botchko botchko@lut.fi](https://reader036.fdocuments.net/reader036/viewer/2022070502/56814c29550346895db92e87/html5/thumbnails/26.jpg)
26
Using a ratio image to enhance road detail (two upper is a multispectral image components)
The third image (lower) is the dividend of the first two
![Page 27: Vladimir Botchko botchko@lut.fi](https://reader036.fdocuments.net/reader036/viewer/2022070502/56814c29550346895db92e87/html5/thumbnails/27.jpg)
27
Color analysis. Color similarity
Brick Ceramic tiles Wooden pieces Car parts
![Page 28: Vladimir Botchko botchko@lut.fi](https://reader036.fdocuments.net/reader036/viewer/2022070502/56814c29550346895db92e87/html5/thumbnails/28.jpg)
28
Color analysis. The Munsell Book of Color contains a set of color patches
http://www.it.lut.fi/research/color/demonstration/demonstration.html
![Page 29: Vladimir Botchko botchko@lut.fi](https://reader036.fdocuments.net/reader036/viewer/2022070502/56814c29550346895db92e87/html5/thumbnails/29.jpg)
29
Color analysis. Metameric spectra. Color is the same at one illumination (left patches) and different at another
illumination (right patches).
![Page 30: Vladimir Botchko botchko@lut.fi](https://reader036.fdocuments.net/reader036/viewer/2022070502/56814c29550346895db92e87/html5/thumbnails/30.jpg)
30
![Page 31: Vladimir Botchko botchko@lut.fi](https://reader036.fdocuments.net/reader036/viewer/2022070502/56814c29550346895db92e87/html5/thumbnails/31.jpg)
31
Statistical Analysis of Natural ImagesUpper curve is mean, lower curve is standard deviation
![Page 32: Vladimir Botchko botchko@lut.fi](https://reader036.fdocuments.net/reader036/viewer/2022070502/56814c29550346895db92e87/html5/thumbnails/32.jpg)
32
http://www.techexpo.com/WWW/opto-knowledge/ for previous picture site.http://stargate.jpl.nasa.gov/lctf/ for this picture site.