Title: Sun to Sensor
1Pre-Columbian cities discovered in the Amazon as
displayed here with Landsat.
2Sensor Basics
3Cameras
Camera Basics
Basic operation of a camera is similar to the
human eye in which an object is projected through
a lens to create a reverse image on the film.
4Film Basics
- Most BW films are sensitive from 0.25 to 0.7 ?ms
- UV through red.
- Most Kodak films have extended red for
penetrating atmospheric haze (up to 0.72 ?m) - BW IR films have sensitivity from 0.25 to 0.9
?ms.
5Film Basics
Hudson River in BW Pan film. Note detail in
water areas.
BW IR film. Note the detail in the vegetation
along the road..
6Film Basics
- Color and CIR films have three layers
- Color sensitive to Blue, green, and red.
- CIR sensitive to green, red, and IR
- Color Film
- yellow filter after yellow layer to block blue
light - CIR film
- Yellow filter in front of all emulsions layers to
block blue light.
7Film Basics
8Basic Spectral Reflectance Curves
Camera systems, as we have covered, acquire in
the visible for color or panchromatic film.
9Basic Spectral Reflectance Curves
Camera systems, as we have covered, acquire in
the visible for color or panchromatic film or
infrared with CIR or IR BW film.
10Film Basics
- BW films have the qualities of Tone and Contrast
for interpretation. - Color films add the qualities of Hue, Chroma, and
Value. - Human eye can see about 60 gray tones, but can
discriminate thousands of colors. - Color imagery can be easier and faster to
interpret than black and white. - Downside
- More expensive than BW.
- Slightly less useful for mapping applications.
11Digital Film Products
- Traditionally, the film has been handled in its
hard-copy format (positives, negatives, and
etc). - To use it in the computer, it has to be scanned
(usually at 600 to 2,400 dpi). - B/W or Panchromatic film is scanned into 1 band
(layer). - Color and Color IR film is scanned into 3 bands
(layers). - Therefore the Color Film is scanned into a
visible blue band, visible green band and a
visible red band. Color IR film is scanned into
a visible green band, a visible red band, and a
NIR band. - The ordering of the bands for scanned film is not
standardized, whereas with most satellite systems
the band order starts with the shortest
wavelength (i.e. Band 1 is visible blue, Band 2
is visible green and etc), sometimes scanned
film may be done in reverse order (Band 1 is
visible red, Band 2 is visible green).
12Digital Film Products
- There are now digital camera systems (no film),
although the products tend to be still grouped as
a color or a color IR film product. - Aerial photography tends to have better spatial
resolution than most satellite imagery (at least
until recently) - But their time coverage is inconsistent being
anytime someone contracted an overflight. - The last full coverage of New Mexico was a
digital photo acquisition in 2005-2006 at a 1m
spatial resolution and now 2009 natural color
film mosaics at 1 m spatial resolution are coming
online. You can get the data at the RGIS home
page (http//rgis.unm.edu/). - The first coverage of most of New Mexico was in
the 1930s.
13Digital Sensor Types
- electro-optical scanner (cross-track or
whisk-broom scanner) a set of sensors for each
bandwidth that record responses as the scanning
mirror passes over an area. They have the
disadvantage of being dependent on a
well-functioning scanning mirror. They have the
advantage of having a wide variety of detectors
available. - Example Landsat, MODIS
- charge-coupled device (CCD, along-track or
push-broom imager) a fixed array of sensors
acquiring a swath of data at one time. It has
the advantage of being able to stay longer over
an area allowing for longer dwell time. It has
the disadvantages of having many more sensors
that can go wrong and the CCDs have only been
developed for a few bandwidths. Example SPOT,
IKONOS
14Digital Sensor Types
- electro-optical scanner (cross-track or
whisk-broom scanner) a set of sensors for each
bandwidth that record responses as the scanning
mirror passes over an area. They have the
disadvantage of being dependent on a
well-functioning scanning mirror. They have the
advantage of having a wide variety of detectors
available. - Example Landsat, MODIS
- charge-coupled device (CCD, along-track or
push-broom imager) a fixed array of sensors
acquiring a swath of data at one time. It has
the advantage of being able to stay longer over
an area allowing for longer dwell time. It has
the disadvantages of having many more sensors
that can go wrong and the CCDs have only been
developed for a few bandwidths. Example SPOT,
IKONOS
15Sensor Basics
- The older type of multi-spectral scanners found
on satellites and airplanes is the whisk-broom or
cross-track scanner.
16Sensor Basics
- Digital cameras and push-broom (along-track)
scanners found on satellites, have the exterior
object is projected on to an array of detectors
sensitive to an electromagnetic energy flux.
17Satellite Orbits
- Geo-stationary Orbits satellites stationed
far enough to rotate at the same speed as Earths
rotation (36,000 km). This allows for constant
monitoring, but poor spatial resolution. - Example weather satellites
- Polar Orbits Satellites stationed closer to
Earth (100s kms), traveling north to south at a
slight incline allowing the Earths rotation to
provide the east to west coverage. This allows
for better spatial resolution, but less than
daily revisit. - Example Landsat, SPOT, MODIS, IKONOS
- Equatorial Orbits Satellites that orbit
around the equator (usually 50?N to 50?S) which
provide multiple revisits in a day although at
different times each day and no polar coverage. - Example TRMM, Space Shuttle, Space Station
18Polar Orbit Landsat Example
19Orbital Path and Row Landsat Example
SWATH WIDTH
20Several Considerations of Satellite Imagery
- Spectral Resolution
- Spatial Resolution
- Radiometric Resolution
- Temporal Resolution revisit time
- Total Coverage swath width, scene size
-
21Spectral Resolution
- Several types of spectral capabilities
- panchromatic (1 band),
- multi-spectral (around 10 or less bands),
- hyper-spectral (hundreds of bands).
22Spectral Resolution
The higher the spectral resolution the better the
ability of distinguishing between different
surface features, but it comes at a cost in
disk space, CPU processing and quite literally
money.
23Spatial Resolution
- Spatial resolution is a function of
- altitude of the sensor,
- optics and sensitivity of the sensor,
- dwell time,
- and amount of energy available in the bandwidth.
24Spatial Resolution
- Some common sensors with one cell spatial
resolution mapped out.
25Radiometric Resolution
1 bit 8 bit
- Once detected by the sensor, the response is
digitally recorded as binary data. The higher
the response, the higher value recorded (this is
as an integer which is known as a DN value
digital number value). - Binary data records the data in a series of bits
with each bit switched to on or off (0 or 1).
The pattern of bit response then is translated by
a computer as to the DN value. - The higher the bit value of the system the more
DN values it can record, the more detail is
available.
26General Satellite Comparison
27Some General Rules about Satellite Imagery
- Spatial resolution and cost are directly
related. - Spatial resolution and scene coverage are
inversely related - High spatial resolution platforms tend to only
have VNIR (Visible and NIR). - Moderate to low spatial resolution platforms can
have MIR and TIR capabilities. - Direct nadir revisit times are longer for
higher spatial resolution Most higher spatial
resolution satellites get around this by having
pointable systems.
28Commercially Available Imagery
- AVHRR
- Landsat
- IRS
- SPOT
- Ikonos/GeoEye
- DigitalGlobe
- Radarsat
- ERS/Envisat
29-AVHRR-Advanced Very High Resolution Radiometer
- Data available from a number of platforms since
the early 1970s. - Multispectral 1.1 km (0.58-0.68 ?ms,
0.725-1.1 ?ms, 3.55-3.93 ?ms, 10.3-11.3 ?ms,
11.5-12.5 ?ms) - Swath width 2,700 kms
- Revisit time twice daily
- Dynamic range 16-bits
- edcwww.cr.usgs.gov/landdaac/1KM/avhrr.sensor.htm
l
30AVHRR Spectral Reflectance Curves
The AVHRR sensors has 5 bands which image the
VNIR and in the TIR. (Band 1 visible red, Band
2 NIR, Band 3 TIR3.55-3.93 ?ms , Band 4
TIR10.3-11.3 ?ms, and Band 5 TIR11.5-12.5 ?ms ).
31Landsat Multi-Spectral Scanner (MSS)
- Landsat 1 launched on July 23, 1972 (operations
ceased in 1978). - Landsat 2 launched on January 22, 1975
(operations ceased in 1982). - Landsat 3 launched on March 5, 1978 (operations
ceased in 1983). - Landsat 4 launched July 16, 1982 (operations
ceased in 1993). - Landsat 5 Launched March 1, 1984 (still
operational). - Multispectral 57 m VNIR, (0.5-0.6 ?ms,
0.6-0.7 ?ms, 0.7-0.8 ?ms, 0.8-1.1 ?ms) - Swath width 185 kms
- Revisit time 18 days for Landsats 1-3, 16
days for Landsats 4 and 5. - Dynamic range 8-bits
- Most images archived and free.
32Landsat Thematic Mapper (TM)
- Landsat 4 launched July 16, 1982 (operations
ceased in 1993). - Landsat 5 Launched March 1, 1984 (still
operational). - Landsat 6 never made orbit.
- Multispectral 30 m VNIR, MIR, 60 m TIR
(0.45-0.52 ?ms, 0.53-0.61 ?ms, 0.63-0.69 ?ms,
0.75-0.9 ?ms, 1.55-1.75 ?ms, 2.09-2.35 ?ms,
10.4-12.5 ?ms) - Swath width 185 kms
- Revisit time 16 days
- Dynamic range 8-bits
- Most images archived and free.
33Landsat Enhanced Thematic Mapper (ETM)
Landsat 7 Launched May, 1999 (scan line
correction error May 31, 2003 has made any data
after this date unusable for the most
part). Panchromatic 15 m (0.52-0.9
?ms) Multispectral 30 m VNIR, MIR, 60 m TIR
(0.45-0.52 ?ms, 0.53-0.61 ?ms, 0.63-0.69 ?ms,
0.75-0.9 ?ms, 1.55-1.75 ?ms, 2.09-2.35 ?ms,
10.4-12.5 ?ms) Swath width 185 kms Revisit
time 16 days Dynamic range 8-bits Most
images archived and free.
34Landsat Spectral Reflectance Curves
Landsat TM and ETM sensors have 7 bands which
image the VNIR, the MIR and in the TIR. (Band 1
visible blue, Band 2- visible green, Band 3
visible red, Band 4 NIR, Band 5 MIR1.55-1.75
?ms , Band 6 TIR10.4-12.5 ?ms, and Band 7
MIR2.09-2.35 ?ms ).
35Landsat Thematic Mapper characteristics
- Landsat 7 Enhanced Thematic Mapper only.
- Order images through http//edcsns17.cr.usgs.gov/E
arthExplorer
3610
Landsat Non-Thermal bands
Bare soil
Spectral Reflectance
Senescent vegetation
Green Vegetation
Wavelength (µm)
37GeoEyes Ikonos
Launched September 24, 1999 Panchromatic -
1m (0.53-0.929 ?ms) Multispectral - 4 m
(0.45-0.52 ?ms, 0.51-0.6 ?ms, 0.63-0.7 ?ms,
0.76-0.85 ?ms ) Swath width 11 kms Revisit
time 3 days Dynamic range 11-bits
www.geoeye.com/CorpSite
38GeoEyes GeoEye-1
Launched September 6, 2008 Panchromatic
0.41 m (0.45-0.8 ?ms) Multispectral 1.65 m
(0.45-0.51 ?ms, 0.51- 0.58 ?ms, 0.655-0.69 ?ms,
0.78-0.92 ?ms ) Swath width 15 kms Revisit
time 3 days Dynamic range 11-bits
www.geoeye.com/CorpSite
39DigitalGlobes Quickbird/Worldview
Launched October, 2001 Panchromatic 0.65
m (0.45-0.9 ?ms) Multispectral 2.6 m
(0.45-0.52 ?ms, 0.52-0.6 ?ms, 0.63-0.69 ?ms,
0.76-0.9 ?ms ) Swath width 16.5 kms Revisit
time 7 days Dynamic range 11-bits
www.digitalglobe.com In September, 2007,
WorldView-1, with a 0.5 m Pan sensor was launched.
40High Spatial Resolution Sensor Characteristics
Digital Globe, GeoEye and IKONOS satellite images
have a Pan that covers the VNIR as well as 4
separate Multi-Spectral bands in the VNIR (Band 1
visible blue, Band 2- visible green, Band 3
visible red, Band 4 NIR).
41High Spatial Resolution Sensor Characteristics
- Expensive - 10s/km2
- Can be ordered geo-corrected or
ortho-corrected. - Only tasked not continually acquired, some
archived.
42Radarsat
- Launched November, 1995
- C-band , HH
- Swath widths of 50-500 km
- Spatial resolutions of 8-100 m
- Incidence angles from 10-59?
- www.rsi.ca
43Radarsat
- Launched November, 1999
- C-band (5.7 cm) HH
- Swath widths of 50-500 km
- Spatial resolutions of 8-100 m
- Incidence angles from 10-59?
- www.rsi.ca
44RSIs Radarsat-2
Launched December 14, 2007 C-band (5.7 cm)
HH, VV, HV, VH Swath widths of 50-500 km
Spatial resolutions of 3-100 m Incidence angles
from 10-59? Revisit daily to once every 2-3
days www.radarsat2.info/
45(No Transcript)