Photogrammetry and Multispectral Remote Sensing

1
Photogrammetry and Multispectral Remote Sensing

• Lecture 3
• September 8, 2004

2
What is Photogrammetry

• Photogrammetry is the art and science of making
  accurate measurements by means of aerial
  photography
• Analog photogrammetry (hard-copy photos)
• Digital photogrammetry (digital images)
• Aerial photographs were the first form of remote
  sensing imagery.
• Differences between photogrammetry and Remote
  Sensing are that photographs are
• Black and white (1 band) or color (blue, green,
  red, and IR)
• Wavelength range of 0.3-1.0 ?m
• Use cameras
• One type of remote sensing imagery

3
Types of vantage points to acquire photographs

• Vertical vantage points
• Low-oblique vantage points
• High-oblique vantage points

4
Vertical Aerial Photography
Goosenecks of the San Juan River in Utah
Jensen, 2000
Most are vertical aerial photography
5
Low-oblique Aerial Photography
Low-oblique photograph of a bridge on the
Congaree River near Columbia, SC.
Jensen, 2000
6
High-oblique Aerial Photography
High-oblique photograph of the grand Coulee Dam
in Washington in 1940
Jensen, 2000
7
Color Science
additive

• Additive primary colors
• Blue, Green, and Red
• Subtractive primary colors (or complementary
  colors)
• Yellow, Magenta, and Cyan
• Filters (subtract or absorb some colors before
  the light reaches the camera)
• Red filter (absorbs green and blue, you can red)
• Yellow (or minus-blue) filter (absorbs blue,
  allows green and red to be transmitted, which is
  yellow)
• Haze filter (absorbs UV)

Subtractive
8
Types of photographs

• Black and white photographs
• Panchromatic (minus-blue filter used to eliminate
  UV and blue wavelengths)
• IR (IR-sensitive film and IR only filter used to
  acquire photographs at 0.7- 1.0 ?m )
• UV (at 0.3-0.4 ?m, low contrast and poor spatial
  resolution due to serious atmospheric scattering)
• Color photographs
• Normal color (Haze filter used to absorb UV and
  create true color 0.4-0.7 ?m, or blue, green,
  red)
• IR color (Yellow filter used to eliminate blue
  and create IR color of 05-1.0 ?m, or green, red,
  IR)
• 4 bands (blue, green, red, and IR)

9
Satellite photographs

• Extensive collections of photographs have been
  acquired from manned and unmanned Earth or
  Mars-orbiting satellites.
• Beginning in 1962, USA acquired photographs of
  moon for Apollo mission
• 1995, USA declassified intelligence satellites
  photographs of Sino-Soviet acquired 1960-1972 at
  2-8 m resolution.
• 2000, Russia launched satellites acquired
  photographs of 2 meter resolution
• 1999, Mars Global Surveyor of NASA acquires Mars
  photographs with 1.2 12 m resolution
• 2003, Mars Express of ESA acquires Mars
  photographs with 2 and 10 m resolution.

10
Flightline of Vertical Aerial Photography
Jensen, 2000
11
Block of Vertical Aerial Photography
Jensen, 2000
12
Block of Vertical Aerial Photography Compiled
into an Uncontrolled Photomosaic
Columbia, SC Original scale 16,000 Focal
length 6 (152.82 mm) March 30, 1993
Jensen, 2000
13
Scale of photographs

• Image size/ real world size
• S ab/AB
• Focal length/ altitude above ground
• S f / H

14
Scale (2)
1 12
S f / (H-h)
0.012/ (6 x 12) 1/6000
Max scale, minimum scale, and average or nominal
scale
15
Orthophotographs and digital orthoimagery

• Photograph after corrected by ground control
  points (x, y, z) or digital elevation model (DEM)
  called orthophotograph, orthophoto, or digital
  orthoimagery.
• Not as photographs, they have different scales in
  different terrain relief, orthophotos have only
  one scale, no distortion, and have true distance,
  angle, and area. Orthophotos can be directly
  input into GIS as basemap or for interpretation.

16
Extraction of Building Infrastructure based on
orthophotographs
17
Orthophotograph draped over a DEM
18
Kevin Hankinson will share his experience in
acquiring aerial photos
19
Multispectral Remote Sensing

• Multispectral remote sensing is defined as the
  collection of reflected, emitted, or
  backscattered energy from an object or area of
  interest in multiple bands of electromagnetic
  spectrum while Hyperspectral remote sensing
  involves data collection in hundreds of bands.
• Instead of cameras and 1 or 4 bands for
  photogrammetry, Remote sensing use detectors that
  are sensitive to hundreds of bands in the
  electromagnetic spectrum. Measurements made by
  detectors are always stored in a digital format.

20
Overview
Jensen, 2000
Energy detected is recorded as an analog
electrical signal
21
Remote Sensing Raster (Matrix) Data Format
Y axis
Jensen, 2000
22

• Discrete Detectors and scanning mirrors
• - MSS, TM, ETM, GOES, AVHRR, SeaWiFS, AMS,
  ATLAS
• Linear Arrays
• - SPOT, IRS, IKONOS, ORBIMAGE, Quickbird, ASTER,
  MISR
• Liner and area arrays
• - AVIRIS, CASI, MODIS, ALI, Hyperion, LAC

Detector configurations breaking ou the spectrum
Jensen, 2000
23
Sabin, 1997
Liner arrays and area arrays
Discrete Detectors and scanning mirrors
Field of View (FOV), Instantaneous Field of View
(IFOV) Dwell time is the time required for the
detector IFOV to sweep across a ground cell. The
longer dwell time allows more energy to impinge
on the detector, which creates a stronger signal.
24
Landsat satellite series
MSS TM ETM 0.5-0.6
0.45-0.52 0.45-0.52 0.6-0.7 0.52-0.60
0.52-0.61 0.7-0.8 0.63-0.69
0.63-0.69 0.8-1.1 0.76-0.90
0.78-0.90 10.4-12.6 1.55-1.75 1.55-1.75
10.4-12.5 10.4-12.5
2.08-2.35 2.09-2.35
0.52-0.90 79m 30
30 240m 120 60
15 6 bits 8
8 103 m/c 99 99 18 days
16 16 919km 705
705 185km 185 185
Jensen, 2000
25
Inclination (99º) of the Landsat Orbit to
Maintain A Sun-synchronous Orbit
MSS 99º TM 98.2º

• Sun-synchronous orbit mean that the orbital plane
  precessed around Earth at the same angular rate
  at which Earth moved around the Sun
• The satellite cross the equator at approximately
  the same local time (930 to 1000 am)

Jensen, 2000
26
Todays Landsat 7 orbits and acquisition
http//landsat7.usgs.gov/pathrows.php
27
TexasView Remote Sensing Consortium
Free Landsat 7 imagery available from TexasView
http//www.texasview.org/pages/archives/html/lands
at.html
28
About the lab setup and a DEMO of ENVI