Visible and NIR

Report
EE/Ae 157a
Week 4: Visible and Near IR
4-1
Topics to be Covered
•
Space Mirrors
– Diffraction limited resolution, Space mirror materials, Mirror coatings,
structural materials
•
Space Detectors
– Photoemissive, Photoconductive, Photovoltaic, CCDs
•
Examples of Systems
– Landsat MSS and TM, SPOT
•
Examples of Image Artifacts
– Line Dropouts, Banding, Line Offsets
•
Analysis Techniques
– Ratio images, Principal components, NDVI, Edge enhancements,
Sharpening, Spectral unmixing, Classification
4-2
Basic Remote Sensing System
Source
Detector
Waves Emitted
Scattering Object
Collecting
Aperture
4-3
Imaging Terms
Field-of-view
Swath
Width
Dwell Time
Cross-Track
Direction
Along-Track
Direction
4-4
Areal Image Plane
Imaging Optics
Types of Imaging Systems
Platform
Movement
Scanning
Mirror
Imaging
Optics
“Point”
Detector
Along-Track Direction
Swath
Width
Cross-Track Direction
(a) Framing Camera
(b) Scanning System
Line Array
Detectors
Imaging Optics
4-5
(c) Pushbroom System
Comparison of Imaging Systems
T ype
Advantage
Disadvantage
Film framing camera
Large image format
High information density
Cartographic accuracy
Transmission of film
Potential image smearing
Electronic framing camera
Broad spectral range
Data in digital format
Good geometric fidelity
Difficulty in getting large arrays
Wide field-of-view optics
Scanning systems
Simple detector
Narrow field-of-view optics
Wide sweep capability
Easy to implement multiple
wavelenghts
Low detector dwell time
Moving parts
Difficult to achieve good
geometric fidelity
Pushbroom imagers
Long dwell time per detector
Good cross-track geometric
fidelity
Wide field-of-view optics
From Elachi,1987
4-6
Basic Telescope
Primary
Secondary
Focal Plane
f  FD
f  focal length
F  focal rat io
D  apert ure diameter
4-7
Diffraction Limited Resolution
Circular Aperture
d  2.44

D

Rayleigh criterion for resolution:   1.22 D
4-8
Telescope Classification
T elescope
Primary Mirror
Secondary Mirror
Cassegrain
Parabola
Hyperbola
Gregorian
Parabola
Ellipse
Ritchey-Chritein
Hyperbola
Hyperbola
Dall-Kirkham
Ellipse
Sphere
Newtonian
Parabola
Flat
Schmidt
Aspheric
Sphere
Scwarzschild
Sphere
Sphere
From Space Remote Sensing Systems: An Introduction, by H.S. Chen, 1985
4-9
Types of Telescopes
Newtonian
Ritchey-Critien
Cassegrain
Schwarzschild
Gregorian
Dall-Kirkham
Schmidt
From Space Remote Sensing Systems: An Introduction, by H.S. Chen, 1985
4-10
Telescope Terms
• Focal Length: Effective length of the light path from
the lens or mirror to the focus point
• Aperture Size: Unobstructed size of the lens or mirror
• Focal plane: The area covered with sensors that
change electromagnetic energy into electrical signals
• Field of View: The angle viewed by the focal plane
• Pixel Field of View: The angle viewed by a single
detector in the focal plane
• Field of Regard: The total angle that a scanning
telescope can image
4-11
Diffraction Limited Resolution
Circular Aperture
d  2.44

D

Rayleigh criterion for resolution:   1.22 D
4-12
Diffraction Limited Resolution
Circular Aperture
Separation < 1.22 /D
Separation = 1.22 /D
Separation > 1.22 /D
4-13
Diffraction Limited Resolution
Effect of Wavelength
4-14
Diffraction Limited Resolution
Aperture Size for Constant Resolution
1000000
Apert ure Size in Meters
100000
10000
1000
100
10
1
0.1
10
1000
100000
10000000
W avelength in Microns
4-15
Diffraction Limited Resolution
Effect of Apodization
NO APODIZATION
GAUSSIAN SIGMA = RADIUS
4-16
Diffraction Limited Resolution
Effect of Apodization
NO APODIZATION
GAUSSIAN SIGMA = RADIUS
4-17
Diffraction Limited Resolution
Effect of Aperture Shape
4-18
Diffraction Limited Resolution
Effect of Surface Errors
NO ERRORS
WAVELENGTH / 10
4-19
Diffraction Limited Resolution
Effect of Surface Errors
NO ERRORS
WAVELENGTH / 10
4-20
Diffraction Limited Resolution
Effect of Surface Errors
NO ERRORS
WAVELENGTH / 6.66
4-21
Improving Angular Resolution Through
Aperture Synthesis
4-22
Improving Angular Resolution Through
Aperture Synthesis
LIGHT ADDED IN PHASE
4-23
Improving Angular Resolution Through
Aperture Synthesis
LIGHT ADDED OUT OF PHASE
4-24
Aperture Synthesis
Effect of Aperture Spacing
Spacing = 4 diameters
Spacing = 8.5 diameters
4-25
Space Mirror Materials
Material
Density
Modulus of Elacticity
(N/cm 2 x 106)
Coefficient of Thermal
Expansion
(1/oC x 10-6)
Fused Silica
2.20
7.0
0.55
ULE
2.21
6.74
0.03
Cer-Vi t
2.50
9.23
0.1
Zerodur
2.52
9.20
0.05
Beryl l i um
1.86
28.0
12.4
Al umi num
2.70
6.9
23.9
Invar
8.0
14.8
1.3
Graphi te Epoxy
1.72
6.89
1.0
From Space Remote Sensing Systems, by H.S. Chen
4-26
Space Mirror Coatings
100
90
80
% Reflect ance
70
60
50
Alumi num
40
Silver
30
Gol d
Copper
20
10
0
0.2
0.4
0.6
0.8
1
W avelength in microns
Adapted From Space Remote Sensing Systems, by H.S. Chen
4-27
Space Structural Materials
Characteristics
Al
Be
Gr/Ep
Gr/Al
Gr/Mg
Light Weight
Fair
Fair
Good
Good
Good
H igh Modulus
Fair
Good
Fair
Good
Good
N o Outgassing
Fair
Fair
Poor
Poor
Poor
Conductivity
Fair
Fair
Poor
Fair
Fair
Cost
Low
Medium
Medium
High
High
From Space Remote Sensing Systems, by H.S. Chen
4-28
Detectors
•
•
•
•
•
•
Electro-optical detectors transforms wave energy into electrical
energy
The two most common types are thermal and quantum detectors
Thermal detectors rely on the increase in temperature in heat
sensitive material due to absorption of incident radiation
Implementations include bolometers and thermocouplers
Thermal detectors are slow, have low sensitivity, and their
response is independent of wavelength
Thermal detectors are not commonly used in modern remote
sensing systems
4-29
Detectors
•
•
•
•
Quantum detectors use the direct interaction of the incident
photons with the detector material, which produces free charge
carriers
They are typically classified into three categories: photoemissive,
photoconductive, and photovoltaic
Quantum detectors have fast response and high sensitivity, but
have a limited spectral response
Quantum detectors are characterized by a parameter
A f
D 
NEP
*
A  Det ect or area
f  Bandwidt h
NEP  Noise equivalent power
4-30
D
*
4-31
Photoemissive Detectors
•
•
•
•
•
•
•
In photoemissive detectors, the incident radiation leads to
electron emission from a photosensitive intercepting surface
The emitted electrons are accelerated and amplified
These detectors are primarily used at shorter wavelengths, since
the incoming photons must have sufficient energy to overcome
the binding energy of the electrons
Cesium has a cut-off wavelength of 0.64 microns
Composites, such as silver-oxygen-cesium have longer
wavelength (1.25 microns) cut-off wavelength
An example of this type of detector is the Photomultiplier tube
(PMT)
Landsat multi-spectral scanner (MSS) used PMT detectors for
three of the four bands
4-32
Photoconductive Detectors
•
•
•
•
•
•
In photoconductive detectors, photons with incident energy
greater than the forbidden band energy gap in the semiconductor
material produces free-charge carriers
This causes the resistance of the photosensitive material to vary
inversely proportional to the number of incident photons
Exciting electrons across the forbidden band requires
substantially less energy than electron emission, and
consequently photoconductive detectors can operate at longer
wavelengths
Back-biased silicon photodiodes operate in the photoconductive
mode
Photodiodes can respond within a few nanoseconds
Landsat MSS band 4 used a photodiode as a detector.
4-33
Photovoltaic Detectors
•
•
•
•
In the case of photovoltaic detectors, the incident energy is
focused on a p-n junction, modifying the electrical properties, such
as the backward bias current
Unbiased silicon photodiodes operate in the photovoltaic mode
Because this mode has no dark current, it has distinct advantages
for low-level dc radiation signals
The photovoltaic response time is typically limited to a few
microseconds
4-34
Detector Landscape
1-10 nm
10-100 nm
0.1-1 um
UV
1-10 um
Vis
NIR
10-100 um
MIR
100-1000 um
> 1 mm
Sub-mm
mmWave
FIR
Commercial
and defense
applications
in comms
and radar
Primarily driven
by space based
astrophysics
Commercial and defense applications in
terrestrial imaging and sensing
• weak infrastructure
• strong technical infrastructure
• limited funding
• synergistic funding
• great science
• strong
technical
infrastructure
• synergistic
funding
SAFIR
TECHNOLOGIES
CMOS
Micro Channel Plate
InGaAs
SC Calorimeter
CCD Calorimeter
CCD
GaN
SIS
Uncooled Bolo
QWIP
HgCdTe
InSb
HEB
Si: As
Schottky
SC Bolometer
Si: Sb
Ge: Ga
InP HEMT
4-35
Charge Coupled Device (CCD) Detectors
•
•
•
•
•
•
CCD devices control the movement of signal electrons by the
application of electric fields
Most CCD devices can operate in either the photoconductive or
the photovoltaic modes
In monolithic CCDs the photon detection and multiplexing are
performed on the same chip. It is best suited for VLSI technology,
and have lower production costs
In hybrid CCDs these operations are performed by two separate
chips. Splitting these operations means that each can be
optimized separately
CCD detectors are easily integrated into arrays
Most modern remote sensing systems use CCD detectors.
Examples include SPOT, MOMS and Galileo
4-36
CCD Readout
4-37
CCD Timing
4-38
Example: Kodak CCDs
Device
Pixels (HxV)
Pixel Size (H x Vµm)
KAF-0261E
512 x 512
20.0 x 20.0
KAF-0401E(/LE)
768 x 512
9.0 x 9.0
KAF-1001E
1024 x 1024
24.0 x 24.0
KAF-1301E(/LE)
1280 x 1024
16.0 x 16.0
KAF-1401E
1320 x 1037
6.8 x 6.8
KAF-1602E(/LE)
1536 x 1024
9.0 x 9.0
KAF-3200E(ME)
2184 x 1472
6.8 x 6.8
KAF-4301E
2084 x 2084
24.0 x 24.0
KAF-6303E(/02LE)
3088 x 2056
9.0 x 9.0
KAF-16801E(/LE)
4096 x 4096
9.0 x 9.0
4-39
Landsat 7 Orbit
Parameter
Value
Orbit Altitude, km
705.3
Orbit Period, min
98.9
Orbit Inclination, deg
98.2
Repeat Cycle, days
16
Orbit Type
Sun synchronous
Image Time
10:00 a.m local time
4-40
Landsat ETM+
Parameter
Value
Telescope Diameter, cm
Telescope Type
f-number
Swath Width, km
Scan Frequency, Hz
Scan Angle, deg
Number of lines per scan
Ground Resolution, m
Bandpass, m
Band 1
Band 2
Band 3
Band 4
Band 5
Band 6
Band 7
Band 8
Quantization Level
40.6
Ritchey-Chritien
6
185
7.0
7.7
16
15/30 / 60
0.45-0.52
0.52-0.60
0.63-0.69
0.76-0.90
1.55-1.75
10.4-12.5
2.08-2.35
0.52-0.90 (15m)
256 (8 bits)
4-41
Landsat TM Optical System
4-42
SPOT
Parameter
Value
Orbit Altitude, km
Orbit Type
Image Time
Swath Width, km
Imager Type
Number of detectors per line
Detector Type
Ground Resolution, m
822
Sun synchronous
10:30 a.m.
60
Pushbroom
6000 / 3000
CCD Arrays
10 / 20
Bandpass, m
Band 1
Band 2
Band 3
Band 4
0.50-0.59
0.61-0.69
0.79-0.90
0.50-0.90
4-43
SPOT vs LANDSAT
4-44
Analysis Techniques
Color Combinations
4-45
Analysis Techniques
Color Combinations
4-46
ASTER Parameters
Subsystem
VNIR
SWIR
TIR
Spectral Range (  m)
Spatial
Resolution
(m)
1
0.52-0.60
15
8
2
0.63-0.69
3N
0.78-0.86
3B
0.78-0.86
4
1.60-1.70
30
8
5
2.145-2.185
6
2.185-2.225
7
2.235-2.285
8
2.295-2.365
9
2.360-2.430
10
8.125-8.475
90
12
11
8.475-8.825
12
8.925-9.275
13
10.25-10.95
14
10.95-11.65
Band
No.
Quantization
Levels
(bits)
4-47
ASTER DATA OF CUPRITE, NV
4-48
ASTER Color Combinations
1-2-3
1-3-6
4-49
Analysis Techniques
Ratio Images
•
•
•
•
•
Ratio images are formed by dividing the data value in one band by
that of another band
Ratio images are used to emphasize differences in spectral
reflectance of materials. For example, vegetation shows a
maximum reflectance in TM Band 4 and a lower reflectance in
band 2. The ratio image 4/2 enhances the vegetation signature
Ratio images minimize the difference in illumination conditions,
and suppress the effects of topography
A disadvantage is that ratio images suppress differences in
albedo; materials with different albedos but similar spectral
properties may not be distinguishable in ratio images
Another disadvantage is that noise is emphasized in ratio images
4-50
Ratio Images
100
90
80
% Reflectance
70
Kaolinite
Alunite
Hematite
Montmorillonite
Jarosite
Goethite
Buddingtonite
Muscovite
60
50
40
30
20
10
2
1
0
0.50
4
6
3
5
1.00
1.50
Wavelength in Microns
2.00
8
7
9
2.50
4-51
Ratio Images: Band 4/7
Highlights presence of clays due to Al-OH
bending mode absorption feature in band 7
4-52
Ratio Images: Band 3/1
Highlights presence of iron oxides
4-53
Ratio Images: Band 4/3
Highlights presence of iron oxides
4-54
Ratio Images: Color Combination
4-55
Analysis Techniques
NDVI
•
The normalized difference vegetation index (NDVI) is defined as
NDVI 
•
•
ni r  re d
ni r  re d
Higher values of NDVI indicate higher concentration of green
vegetation
NDVI maps are typically calculated using biweekly combinations
of images to reduce the effects of cloud cover
4-56
Analysis Techniques
NDVI
4-57
Analysis Techniques
Intensity/Hue/Saturation Transformation
4-58
Analysis Techniques
Intensity/Hue/Saturation Transformation
4-59
Analysis Techniques
Sensor Combinations
4-60
Analysis Techniques
Principal Components
•
•
•
•
•
•
Typically, images from individual bands are highly correlated on a
pixel by pixel basis
The principal component transformation arranges images in order
of the amount of variance in the data across the image
This mathematical transformation is similar to calculating the
eigenvalues and eigenvectors of the image on a pixel by pixel
basis
Most of the variance is typically in the first few principal
components, with the last few dominated by noise
The first PC image is typically dominated by topographic effects
By displaying three PC images as red, green and blue, spectral
variations are typically enhanced
4-61
Analysis Techniques
Principal Components
4-62
Analysis Techniques
Principal Components
4-63
Analysis Techniques
Principal Components
1–2-3
PC1 – PC2 – PC3
4-64
Analysis Techniques
Edge Enhancements
•
•
•
•
•
Edge enhancement filters are used to enhance linear features in
images
Geologists use linear features to map faults, while geographers
use linear features to identify man-made structures such as roads
Edges can be enhanced using non-directional or directional filters
An example of a non-directional filter is the Laplace kernel
0
-1
0
-1
4
-1
0
-1
0
Directional edge enhances are used to identify linear features in
specific directions:
0
0
0
0
-2
0
-2
0
0
0
0
-2
-2
4
-2
0
4
0
0
4
0
0
4
0
0
0
0
0
-2
0
0
0
-2
-2
0
0
4-65
Analysis Techniques
Edge Enhancements
4-66
Analysis Techniques
Supervised Classification
4-67
Analysis Techniques
Unsupervised Classification
4-68
Spectral Unmixing
4-69
Spectral Unmixing
4-70

similar documents