The document discusses satellite image geometry and specifications of various satellites like GeoEye-1 and GeoEye-2, detailing their capabilities in capturing and processing high-resolution imagery. It covers technical aspects such as ground sample distance, azimuth, and elevation angles, and explains different image products including orthophotos and stereo images. Additionally, it highlights applications of satellite imagery in mapping, environmental monitoring, and photogrammetry.

1. Satellite Image
Geometry
Gene Dial
GIS in the Rockies
2012-09-21
1

2. Topics
› Background
› Satellite Orbits
› Satellite Image Geometry
› Stereo Image Geometry
› Satellite Image Products
Niagara Falls
2

3. Background

4. Remote Sensing—Then & Now!
Mys Shmidta Air Field, Soviet Union GeoEye-1 Half Meter Imagery
Collected August 18th, 1960 Kutztown University – Collected Oct. 6, 2008

5. GeoEye-2 Technical Specs at a Glance
System Specificatio Performance
n Power Solar Array (5)
Satellite Bus Size 2.3 m x 5.3 m Control
Data Storage Unit (2)
Weight 2100 kg dry, 2500 kg wet Flight
Unit
Payload Aperture 1.1 meter aperture Processor
Focal Length 16 meter focal length Battery (2)
Dynamic Range 11 bit dynamic range with TDI
GSD 34 cm Pan, 1.32m MSI
Swath 14.5 km
Attitude Actuators Honeywell M-95 CMGs Control
CMG Electronics (4)
Moment Gyros
Control (4)
System Sensors Goodrich GR-1004 Star Trackers Star Tracker (2)
SIRU Inertial Reference Units Focal Plane
PL Electronics
Monarch GPS Receiver Radiator (2)
Radiator
Payload
Minimum Agility Acceleration - 1.0 degree/sec2
X-band High
Electronics (2)
Max Slew Rate – 2.7 degree/sec
Gain Antenna Narrowband
Two Axis Gimbal Antenna (2)
Data Data Recorder 3.2 Terabit High Speed Storage Unit
Handling & Sun Sensor (2) GPS Antenna (2)
Communications Wideband DL 800 Mbps Dual Pole, X band
Tlmy DL 128 Kbps, X band
Command UL 64 Kbps, S band
See http://www.youtube.com/watch?v=lnv6cDiBF9o
5

6. Satellite Imagery

7. Satellite Orbits

8. 8

9. GeoEye-1 IKONOS
GeoEye-2
9

10. GeoEye Constellation GeoEye Constellation
GeoEye Constellation
High Resolution Images
Move Beyond Mapping
› Frequent Access
‒ Mean Time to Access < 1 day
› Long Duration Accesses
‒ Average Access Time ~ 1 min/day
› High Resolution Access
‒ GeoEye-1: 41 cm Nadir GSD
‒ GeoEye-2: 34 cm Nadir GSD
‒ True 50 cm products
› Huge Collection Capacity
‒ IKONOS: 240,000 sqkm/day
‒ GeoEye-1: 350,000 sqkm/day
‒ GeoEye-2: 600,000 sqkm/day
Ground Swaths
Ground Swaths
10 GE1 = Green IK = Yellow GE2 = Blue
GE1 = Green IK = Yellow GE2 = Blue

11. Satellite Image Geometry

12. Ground Sample Distance (GSD)
H
GSD
W
GSD
› Source image pixels are rectangular, W x H in size
› GSD = sqrt(W x H)
› A square pixel of GSD x GSD size has the same area as W x H
› Product images may be resampled to a different GSD

13. Satellite Satellite
Field of View imaging at
nadir
imaging
off nadir
› Field of View (FOV) is angle from
one edge of an image to the
other.
› All rays of a high-resolution
satellite image are at about the Satellite
Field of View
same angle.
Camera FOV
Aerial 90°
IKONOS 0.95°
GeoEye-1 1.28°
GeoEye-2 1.22° Aerial Camera Aerial
Field of View camera
13

14. 14

15. Scan Azimuth
› Scan Azimuth
‒ Describes scan direction or motion of aim
0° = North
point on ground
‒ North-to-South
 Scan Azimuth = 180°
‒ South-to-North
 Scan Azimuth = 0° 270° = West 90° = East
180° = South
West to East
Scan azimuth 90°
r o N o h uo S
h u mz A nac S
East to West
t t
Scan azimuth 270°
t i
15

16. Line of Sight (LOS)
› The Line of Sight (LOS) is the
direction that the camera is
imaging.
› A Line of Sight direction can be
described by azimuth and
elevation angles.
16

17. Satellite Collection Geometry
17

18. Azimuth
› Azimuth angle
‒ Measured in the horizontal plane 0° = North
at the target
‒ Angle from north proceeding
clockwise to the projection of the
line of sight into the horizontal 270° = West 90° = East
plane.
‒ Example: 90° azimuth means the
satellite is East of the target
180° = South
when the image is taken.
18

19. Collection Azimuth
› North Up View
19

20. Collection Azimuth
› View from sensor perspective
GE1 Image acquired at 53.5° collection azimuth rotated
180° - 53.5° CW on right to view from sensor perspective.
20

21. Elevation
› Elevation angle
‒ Measured at target
‒ Angle from horizontal plane up to
line of sight.
› Alternatives
Elevation angle
‒ Incidence or Zenith angle
‒ Off-Nadir or Obliquity angle
21

22. Computing height from shadows & layover
DV DV
DH
DH
DV = DH * tan(EL)
22

23. Example: Republic Plaza (Singapore)
Image collected at 67°
elevation angle
Layover measured at 116
m
Height calculation
H = 116 m * tan(67°)
= 273 m
Actual height
280 m
23

24. Elevation angle and terrain displacement
Zenith Sensor
DH = DV / tan(EL)
EL
Earth
DV
DEM
DH
› EL = elevation angle
› DV = vertical distance
› DH = horizontal distance
24

25. Incidence, Elevation, & Off-Nadir Angles
› EL = Elevation = angle at target from horizontal to sensor.
› IN = Incidence = angle at target from zenith to sensor.
› OB = Obliquity = angle at sensor from nadir to target (off-nadir angle)
› IN + EL = 90°
 R Cos( EL) 
› Obliquity is related to elevation by trig formula: OB = ArcSin  e
 (H + R )  
‒ Re radius of earth ~ 6371 km
=
 o e 
‒ Ho = orbit height ~ 681 km
25

26. Incidence, Elevation, & Off-Nadir Angles
26

27. Stereo Geometry Orbit Track
About one minute of orbit time
between left and right image of
a stereo pair.
Ground Track
Convergence
Angle
EL2
EL1
About two seconds of orbit time
AOI to scan a 15 km by 15 km stereo
scene. Longer scans are possible.
A 100 km long stereo pair takes
about 20 seconds to scan.
27

28. Field of Regard (FOR)
› Field of Regard: Angle Range that Camera can Image by
rotating
› Satellite Field of Regard > 90°.
› Field of View can be anywhere within the Field of Regard
28

29. Field of Regard vs. Elevation Angle
› Wider Field of Regard at lower elevation angle
› Wider Field of Regard from higher orbits
29

30. Field of Regard
GSD vs. Cross-Track Distance
1
0.9
0.8
0.7 60° Elevation
60° Elevation
Angle
GSD, m
Angle
0.6
0.5
0.4 IKONOS
GE1
GE2
0.3
0 100 200 300 400 500 600 700 800 900 1000
Cross-Track Distance, km
RunSatComparison
30

31. Revist Time (time between satellite accesses)
3-day revisit at 40° N
at 60° elevation angle
› Shorter revisit time at lower elevation angle & higher latitude
31

32. Revisit Time
› More frequent revisits at
15
high latitudes because the
orbits converge near the
14 7
5
13
poles.
13 6
› Ground stations are located
at high latitudes can contact
12
4
the satellite nerly every
11
10 9
orbital revolution.
3 15
1
2
32

33. Pan-MSI Alignment
› Each MSI pixel covers 4x4 Pan pixels
› 4 multispectral (MSI) bands
› 1 panchromatic (PAN) band
› Simultaneous PAN/MSI collection
› 11-bit resolution
33

34. CIR RGB

35. 4-meter RGB Multispectral 1-meter RGB Pan-Sharpened
Color
enhances
interpretation
for human
visual
perception
1-meter Panchromatic
4-meter CIR Multispectral 1-meter CIR Pan-Sharpened
35

36. Camera Models
36

37. Rational Polynomial Coefficient (RPC) Camera Models
› RPC Camera Models
‒ Generic mathematical model mapping
ground to image coordinates.
‒ Sensor software fits coefficients to
physical camera model of image.
• Sensors
‒ GeoEye, Ikonos, QB, WV, Cartosat …
• Application Software
‒ ERDAS, BAE, PCI, ZI, …
› Applications
‒ Block adjust images with ground control
to improve accuracy.
‒ Orthorectification
‒ Stereo extraction
The mathematics of satellite imagery is
The mathematics of satellite imagery is ‒ Photogrammetry
complicated, but RPC models are simple
complicated, but RPC models are simple

38. Satellite Image Product Geometry

39. Product Geometry
Product Rectification Projection Image Model
Physical (attitude,
Basic Synthetic Array Satellite Scan Path ephemeris & camera
calibrations)
Geo Constant height Map RPC
Ortho DEM Map Ortho
Stereo Constant height Path, Map, or Epi-polar RPC or Physical
Convergence
angle
Elevation
angle
39

40. BASIC
› Photogrammetric
Applications
› Satellite Projected
› Physical Camera Model
‒ High Accuracy
› RPC Camera Model
‒ Rapid Positioning
IKONOS image of the moon (BASIC product)

41. GEO Tsangpo River Basin, Tibet
› Visual Interpretation
‒ Situational awareness
‒ Intelligence
‒ Media
› Photogrammetry
‒ Block adjust with other imagery
or GCP to improve accuracy.
‒ Orthorectify with DEM to
correct for terrain
displacement
› Map Projected
› RPC Camera Model
‒ High Accuracy

42. E BASIC and GEO Products
N
W E
Geo S
S East to West Scan
N
North up Map Projected
BASIC GEO
RPC Model  
Physical Model
 
Projection
Satellite Map
BASIC W
East to West Scan
Satellite Projected

43. Ortho
› Applications
‒ Feature Extraction
‒ GIS Map Base
› Terrain Corrected
› Map Projected
› Mosaics Available
Frankfurt Airport, Germany

44. Georectified or Orthorectified?
› Georectified
Constant Height Line of Sight Topographic
‒ Terrain displace- Surface
ment errors
‒ Quick,
Low cost
› Orthorectified Topo-
graphic
‒ DEM corrects for
Surface
terrain
Ortho-
displacement rectified
Image
‒ Accuracy for
mapping
44

45. Geospatial eXploitation Products™
What is an Orthophoto?
• An orthophoto is an image that Camera
has had all distortion due to Original Image
camera obliquity, terrain relief,
and features removed.
• The SOCET GXP Ortho Manager
converts one or more original
images into an orthophoto by
transforming the pixels to their Orthophoto
proper position according to the
given sensor, terrain, and feature
information.
• In the final product all points in
the image appear as if the DTM
observer were looking down from
nadir position.
March, 2009

46. Orthorectified

47. Georectified

48. Stereo
› Attributes
‒ High resolution
‒ Color
‒ 3-dimensional
› Applications
‒ DEM extraction
‒ 3D feature extraction
‒ Geomorphic
visualization
Stereo RPC Physical
Projection Model Model
Satellite  
Map  
Epi-Polar  
48

49. Stereo Image Of Downtown Denver, Colorado

50. Thank You!
Big Bear Glacier, Alaska