Industrial Training Report- AKTU Industrial Training Report
M4L1.ppt
1. D. Nagesh Kumar, IISc
Remote Sensing: M4L1
Digital Image Processing
Image Enhancement
(i) Concept of Color and
Color Composites
1
2. Objectives
Color Fundamentals
Chromaticity Diagram
Color Models
Color Composites
Remote Sensing: M4L1 D. Nagesh Kumar, IISc
2
3. Color Fundamentals
In eyes, cones are responsible for
color perception.
Of the 6-7 million cones of human
eye:
65% are sensitive to red light
33% to green light
2% to blue light
These form the primary colors
Figure : Spectral response curves for each cone type.
The peaks for each curve are at 440nm (blue),
545nm (green) and 580nm (red).
Remote Sensing: M4L1 D. Nagesh Kumar, IISc
3
4. Color Fundamentals…
Primary Additive colours
: Red, Green and Blue (RGB)
Complementary Colours
: Cyan, Magenta and Yellow
Remote Sensing: M4L1 D. Nagesh Kumar, IISc
4
5. If three primary colors are superimposed in unequal amounts, then number of
colors are produced
If three primary colors are superimposed in equal amounts, then greys ranging
from black to white are produced.
If white light is passed through a color filter, it is possible to subtract one of the
primary colors
Remote Sensing: M4L1
Additive process Subtractive process
D. Nagesh Kumar, IISc
Color Fundamentals- Natural
Color Photography
5
6. Light
Chromatic
Light has a dominant set
of frequencies
Achromatic
Light has no color. Its only
attribute is its intensity
Color models
Standard means to specify colors by defining a 3D coordinate system.
The sub space will contain all possible color combinations within a particular
model
Eg: RBG, CMY, IHS.
Remote Sensing: M4L1 D. Nagesh Kumar, IISc
Color Fundamentals…
6
7. Chromaticity Diagram
It is often convenient to work in a
2D color space.
Chromaticity diagrams show color
composites as a function of x (red),
y (green) and z (which is 1-x-y).
Devices such as colorimeter
measures color using numbers
derived from CIE values
Remote Sensing: M4L1 D. Nagesh Kumar, IISc
7
8. Remote Sensing: M4L1
Primary colors of
red, green and blue are
used within a cartesian
coordinate system
A unit cube is shown
with the underlying
assumption that all
colors are normalized.
Color Space - RGB
D. Nagesh Kumar, IISc
8
9. Color Space - CMY
Cyan (C), Magenta (M) and Yellow
(Y) comprise the secondary colors of
light.
This color space is generally used to
generate hardcopy output.
Remote Sensing: M4L1 D. Nagesh Kumar, IISc
9
10. Remote Sensing: M4L1
Humans define color in
terms of its intensity (I),
hue (H) and saturation (S).
Intensity: Variations in
brightness
Hue: Dominant wavelength
of color
Saturation: Purity of color
D. Nagesh Kumar, IISc
Color Space-IHS
10
11. Remote Sensing: M4L1
Multispectral
Involves simultaneously obtaining images on the same scene at
different wavelengths
Four: Blue, Green, Red and NIR parts of EMR
Multispectral imaging allows the examination of single band
images
• Natural and False colour composites can be produced
False Colour
True colour composite (TCC):
D. Nagesh Kumar, IISc
Photographic Remote Sensing
11
12. Remote Sensing: M4L1
False Colour
True Colour Composite (TCC)
Red band – Red; Green band – Green; Blue band – Blue
False Colour Composite (FCC)
Any other combination of colours
E.g., Blue band – Red; Red band – Green; Green band – Blue
E.g., Blue band – Red; Red band – Green; NIR band – Blue
Standard False Colour Composite (FCC)
E.g., NIR band – Red; Red band – Green; Green band – Blue
In IRS: Band 4 – Red; Band 3 – Green; Band 2 – Blue
D. Nagesh Kumar, IISc
Color Composites
12
13. Remote Sensing: M4L1
Landsat TM
Average Orbital Height: 700 km (440 Miles)
Spatial Resolution: 30 m, except band 6 which is 90 m
Records Data in 7 Wavelength Intervals (bands)
1.Visible Blue (0.45 to 0.52 microns)
2.Visible Green (0.52 to 0.60 microns)
3.Visible Red (0.63 to 0.69 microns)
4.Near Infrared (0.76 to 0.90 microns)
5.Mid Infrared (1.55 to 1.75 microns)
6.Thermal Infrared (10.4 to 12.5 microns)
7.Mid Infrared (2.08 to 2.35 microns)
Bands 1,2,3,4,5, and 7 record reflected energy
Band 6 records emitted thermal (heat) energy
Satellite Images of the Keweenaw Peninsula, USA
D. Nagesh Kumar, IISc
Color Composites
13
14. Optimum Index Factor
When satellites like Thematic Mapper (TM) are capable of
generating more than one color composite, OIF enables to select
the best combination.
OIF is given by the expression:
Where denotes the standard deviation for band and denotes the absolute
value of the correlation coefficient between any two of the three bands which are being
evaluated.
3
1
3
1
)
(
J
J
K
K
R
Abs
S
OIF
K
S K J
R
Remote Sensing: M4L1 D. Nagesh Kumar, IISc
14
