Introduction to Remote Sensing

1
Introduction to Remote Sensing

• John H. Porter

2
Roadmap

• What is Remote Sensing?
• What can we use it to do?
• The Basics of Remote Sensing
• The electromagnetic spectrum
• Representation of remotely-sensed data

3
What is Remote Sensing? Thanks to Wim Bakker -
http//www.itc.nl/bakker/ for assembling the list

• F.F. Sabins in his book "Remote sensing
  principles and interpretation" defines it as
  follows "Remote Sensing is the science of
  acquiring, processing and interpreting images
  that record the interaction between
  electromagnetic energy and matter."
• The United Nations in their annex Principles
  Relating to Remote Sensing of the Earth from
  Space defines it as "The term Remote Sensing
  means the sensing of the Earth's surface from
  space by making use of the properties of
  electromagnetic waves emitted, reflected or
  diffracted by the sensed objects, for the purpose
  of improving natural resources management, land
  use and the protection of the environment."

4
Remote Sensing Definitions

• Lillesand and Kiefer in their book "Remote
  Sensing and Image Interpretation" even define it
  as an art "Remote Sensing is the science and art
  of obtaining information about an object, area,
  or phenomenon through the analysis of data
  acquired by a device that is not in contact with
  the object, area, or phenomenon under
  investigation."
• Charles Elachi in "Introduction to the Physics
  and Techniques of Remote Sensing" "Remote
  Sensing is defined as the acquisition of
  information about an object without being in
  physical contact with it."
• The CCRS had this in their Remote Sensing
  Glossary Remote Sensing (RS) "Group of
  techniques for collecting image or other forms of
  data about an object from measurements made at a
  distance from the object, and the processing and
  analysis of the data."

5
Remote Sensing Definitions

• Common Elements
• Lack of physical contact (remote)
• Development of information (sensing)
• Differences
• restricted to electromagnetic spectrum (or not)
• Objects vs. interactions vs. areas
• Science vs. Art
• Restricted uses (e.g., land use) or not

6
Sources of GIS Data

• Where does the data we use in GIS really come
  from?
• Field Surveys
• surveys using transits and (now) GPS
• Manual Interpretation of Aerial Photos
• Expert digitizes features off of photo using
  digitizing tablet or on-screen digitizing
• Automated Processing of Digital Imagery
• Usually used for land cover mapping

7
Salt marshes are an important, and vulnerable,
part of coastal ecosystems they are
particularly sensitive to sea level changes
here 3.5 mm/yr. How have they changed in the last
40 years?
8
(No Transcript)
9
(No Transcript)
10
(No Transcript)
11
(No Transcript)
12
(No Transcript)
13
A Brief History of Aerial Remote Sensing

• Balloons
• 1858 Paris First photographs taken from a
  balloon by Felix Tournachon
• 1861 - Real-time remote sensing (people with
  telescopes and telegraph sets) was used by the
  Union during the U.S. Civil War by Thaddeus Lowe

Thanks to Compton Tucker and Joseph Burke, UMD
http//www.geog.umd.edu/webspinner/burke/372/lect
ure7.pdf T. Benjal, UMBC http//umbc7.umbc.edu/
tbenja1/santabar/vol1/lec1/1-2.html
14
Brief History

• Rockets
• 1897 - Alfred Nobel
• 1904 - Alfred Maul
• Pigeons
• 1908 - Julius Neubronner
• Kites
• 1906 - G.R. Lawrence
• SF Earthquake

15
Brief History

• Airplanes
• 1903 Wright Brothers first flight
• 1909 W. Wright takes first aerial photo from an
  airplane
• 1915 - J.T.C. More Brabazon first camera
  designed for airplane use
• 1930s routine use of aerial photos for mapping
  purposes

16
Brief History

• Satellites
• 1957 -The era of satellites began with the launch
  of Sputnik
• 1960 - The first serious US photo satellite
  program was Corona film dropped from space
• 1972 Landsat 1 - non-classified electronic
  imagery

17
Filling the GapsVirginia Coast Reserve LTER
Imagery
Resolution 80m 30m 10m 1-10 m 1-5 m
18
Electromagnetic Spectrum

• The electromagnetic spectrum runs from radio to
  visible light, to gamma-rays
• Electromagnetic radiation is characterized by
• Frequency how many cycles in 1 second?
• Wavelength how much distance is traversed in
  one cycle?
• Energy how much energy does a photon carry?
• High frequencies and high energies go together
  and are inversely related to wavelength

19
Electromagnetic Spectrum
20
Absorption Reflection

• The radiation we receive from an object is a
  function of its absorption and reflection, as
  well as the spectrum of the radiation source
  (which is, in turn dictated by its composition
  and temperature).

21
Sample Spectrum from Jupiter
22

• Different substances absorb different spectra of
  electromagnetic energy
• Here is an example of substances in the earths
  atmosphere

23
Light Absorption

• The curve below has the net absorption for the
  atmosphere
• Which wavelengths are going to be best for
  examining features on the ground from space?

24
Spectra of Plants

• Important chemicals in plants also have distinct
  absorption spectra

25
Remote Sensing

• For most remote sensing our focus is on what is
  REFLECTED
• Some sensors in the thermal infrared also deal
  with what is EMITTED
• Aerial photography film is formulated to respond
  to light in specific portions of the spectrum
  particularly in the near infrared
• Satellite imagers have 3 to 200 particular bands
  where they collect data

26
Digital Representation of Images

• To understand how image processing works, it is
  first necessary to understand how digital images
  are represented in the computer
• Images are stored in RASTER form Each pixel takes
  on a radiometric intensity or brightness
  value
• Pixel values typically vary between 0 and 255
  (the maximum value one byte of data can take)

27
Pixels in the raster can be seen when part of the
image is enlarged
UVA Grounds
Pixel value255

• Each pixel has numerical value that represents
  the brightness of that pixel
• Pixels with High values are Bright
• Pixels with Low values are Dim

Pixel value65
28
Note the way values are displayed can be varied.
Below are two versions of the same image. In one
high numerical values are bright, in the other
dim. Both are based on the same raster data
values.
High values are White Low values are dark
High values are Dark Low values are White
29
Color Images

• Color images are differentiated from black and
  white images because they have more than one
  band
• Bands in a color image typically represent the
  brightness in different parts of the spectrum
• For example, one band may depict the brightness
  in the blue part of the spectrum while another
  band represents brightness in the infrared band
• When bands are combined, we can display color
  images

30
Image Display
Brightness in Band 1 shown in RED
Combined image
Band 2 shown in GREEN
When the three colors are mixed, we see a
true-color image
Band 3 shown in BLUE
31
Again, note that we can alter which colors are
used to display which bands. Here we have shifted
two of the bands around. The underlying data
stays the same (band1 is still band1).
Band1 Blue Band2 Green Band3 Red
Band1 Red Band2 Green Band3 Blue
32
Multispectral Imagery

• Images with more than one band are referred to as
  multispectral
• Although you can only display 3 bands at a time
  (only have red, green and blue guns in your
  monitor), images may have many more bands
• Most multispectral satellite images have 3-10
  bands
• Some images can have up to 200 or more bands and
  are referred to as hyperspectral

33
Image Displays dont all need to be of a single
image

• One way of doing a change analysis is to use the
  same band from images taken at different TIMES
• Thus instead of multi-spectral, we have
  multi-temporal displays

34
Changes on Hog Island 1964-1980
35
Pixel Values

• In our input images the individual pixel has
  values that
• Are continuous (usually over integer values
  between 0 and 255)
• Can be expressed as a vector or comma separated
  numbers
• E.g. if for an individual pixel band 1 20, band
  230 and band 3190, we could express that as the
  vector (20,30,190)

36
Characteristics of Images

• Extent what is the land area that is
  represented in the image? E.g., 120x60 km
• Spatial Resolution what are the dimensions of a
  pixel? E.g., 2 m on a side
• Note a high resolution image has a small pixel
  size
• Sometimes spatial resolution is particular to
  specific bands in an image
• Radiometric Resolution what range of brightness
  values can be represented?
• By far the most common is 8-bit (256 unique
  values)
• Some images contain 12-bit (4096) or even 16-bit
  (65,536) distinguishable levels

37
Remote Sensing to GIS
Raw Image
38
Remote Sensing to GIS
Classification
Raw Image
Sometimes steps are omitted or reordered,
depending on the purpose of the analysis
Raster GIS Data Layer
39
Example Thematic Mapper

• The Thematic Mapper is a satellite sensor used in
  recent LANDSAT satellites
• It captures 7 bands
• 30 m spatial resolution for bands 1-5 7, 120 m
  in band 6
• Extent

40
Thematic Mapper Spectral Bands

• Wavelengths in micrometers
• Band 1 0.45-0.52
• Band 2 0.52-0.60
• Band 3 0.63-0.69
• Band 4 0.76-0.90
• Band 5 1.55-1.75
• Band 6 10.40-12.50
• Band 7 2.08-2.35

41
(No Transcript)
42
Thematic Mapper

• Thematic Mapper Band 1 - Blue

43
Band 6

• Thematic Mapper Band 6 Thermal Infrared

44
Band 1 vs Band 6
45
Lecture Sources

• History of Remote Sensing
• Compton Tucker and Joseph Burke, UMD
• http//www.geog.umd.edu/webspinner/burke/372/lectu
  re7.pdf
• T. Benjal, UMBC http//umbc7.umbc.edu/tbenja1/san
  tabar/vol1/lec1/1-2.html
• Electromagnetic Spectrum
• NASA http//imagine.gsfc.nasa.gov/docs/science/kno
  w_l1/emspectrum.html
• Lawrence Berkeley Lab
• http//www.lbl.gov/MicroWorlds/ALSTool/EMSpec/EMS
  pec2.html

46
Lecture Sources

• http//spaceguard.ias.rm.cnr.it/NScience/neo/dicti
  onary/abs-refl.htm
• http//rain.atmos.colostate.edu/AT622_section10.pd
  f
• http//www.biologie.uni-hamburg.de/b-online/e24/3.
  htm
• For more info on the spectrum look at
  http//physics.ucsd.edu/students/courses/fall2002/
  physics9/p9notes/Chapter6.pdf

47
Lecture Sources

• Images
• N.M. Short http//rst.gsfc.nasa.gov/Sect1/Sect1_3.
  html
• USGS
• http//edc.usgs.gov/glis/hyper/guide/landsat_tm
• http//www.satelliteimpressions.com/landsat.html

48
Introduction to Classification of Remotely Sensed
Imagery

• John Porter

49
Automated Image Processing

• This lecture will focus on automated image
  processing
• Unsupervised Classification
• Supervised Classification
• Tools for Image Processing

50
Classification

• Classification is the process of taking the
  brightness values associated with each pixel
  and using them into assign a class to the
  corresponding output pixels
• Output pixels are NOT continuous. Instead they
  are discrete values that represent a class or
  category (e.g. land cover classes)
• For example, we might decide that pixels with
  the value (20,30,190) (in band 1 thru 3 ,
  respectively) indicate that the output pixel
  should be assigned a value of 10
  corresponding to class 10forest

51
Classification

• The trick in classification is to come up with
  rules that will allow us to translate image
  values (e.g., 10,20,190) into classes (forest,
  grass, water, marsh, urban)
• Usually the land cover classes are associated
  with numerical codes e.g., 1forest, 2grass,
  3water, 4urban

52
Classification Methods

• There are two fundamental approaches to
  classification
• Unsupervised
• The computer selects classes based on clustering
  of brightness values
• Supervised
• You specify the classes to be used and provide
  signatures for each class

53
Unsupervised Classification

• Unsupervised classification refers to a variety
  of different techniques that share some features
  in common
• They use statistical clustering techniques to
  decide which pixels should be grouped together
• With luck, these clusters of pixels will
  correspond to land cover classes
• But the correspondence may not be 11
• One land cover class may be represented by more
  than one cluster (easily fixed by recoding)
• One cluster may represent more than one land
  cover type (not easily fixed may need to
  specify more clusters)

54
Example

• The Image Analyst extension in ArcView uses the
  ISODATA unsupervised clustering technique where
  all you need to specify is the number of desired
  classes

55
Each color represents a different cluster
pixels that may correspond to the land cover
classes you are interested in
56
Recoding

• Following an unsupervised classification, you
  need to go through and assign meaning to each
  of the classes (e.g., class 1 water)
• You can use editing functions to set multiple
  classes that represent the same land cover type
  to a common value
• E.g., if both class 1 and class 5 are water
  (albeit, deep water vs. shallow water), you may
  want to edit all the 5s and change them to 1s

57
Supervised Classification

• In supervised classification you help the
  computer to select signatures that represent
  each land cover class
• Signatures are statistical descriptions of the
  brightness values of a given land cover type
  (e.g., the mean band 1 value, the mean band 2
  value etc.)
• You select signatures using a tool that provides
  seed values

58
Used the seed tool and clicked here. The
highlighted marsh area was similar in color and
connected to the point so it was added to the
data used to calculate the signature.
59
The Find Like Areas command highlighted all the
other areas that have a similar color (i.e.
similar spectral values) with skill they will
all be Marsh
60
Classification
Added Water

• The process is repeated to add new classes
• Due to color variations within a class, often
  multiple signatures will be needed to capture a
  single cover class
• The classes can be recoded and lumped (using
  basic editing functions) so that they correspond
  to the desired classes.

• But not all water was captured by the class-
  requiring additional signatures

61
Classification

• Once you have a group of signatures defined, you
  can classify your image. There are several
  methods for doing this
• Paralleliped
• Mahalanobis Distance
• Maximum Likelihood
• And others.

62
Paralleliped Classification

• use maximum and minimum values on each individual
  band (as derived from each signature) to decide
  which pixels fall within a given class
• Advantages
• Fast
• Can be used one signature at a time
• Used by ARCVIEW Image Analyst
• Disadvantages
• Does poorer job separating classes
• Uses only a fraction of the information contained
  in the signature data

63
Maximum Likelihood

• Uses statistical techniques to decide which class
  a pixel falls into
• Advantages
• Uses full signature information (mean, variation
  inter-band covariation) to tease apart similar
  classes
• Disadvantages
• More computer intensive
• Not available in ARCVIEW

64
Graphical Description
Pixel dim on band 1, but bright on band 2

• We can use a graph to display where pixels fall
  on two bands at once

65
Graphical Display

• Here is a sample display where Marsh is dark,
  beach is light on both bands and water is bright
  on one band (presumably blue) and dim on the other

Water
Beach
Marsh
66
Paralleliped
Here, paralleliped would work well

• Paralleliped classification uses boxes based on
  statistical measures of the range (e.g. maximum
  and minimum values)

Max on band 1
Min. on band 1
67
Paralleliped
Areas of overlap lead to uncertain results

• However, sometimes paralleliped may do a poor job
  separating classes

255
Band 2
0
255
0
Band 1
68
Maximum Likelihood

• Maximum likelihood uses seed statistics to define
  ellipses for each class

69
Maximum Likelihood

• Ellipses make it easier to separate similar
  classes

255
Band 2
0
255
0
Band 1
70
Software for Image Classification

• ArcView Image Analyst
• Provides basic image classification capabilities
• Unsupervised ISODATA method
• Supervised Paralleliped only
• Also supports georeferencing and some image
  processing (e.g., sharpen edges)
• Relatively easy to use

71
Software

• ERDAS Imagine
• Full suite of advanced image processing and
  classification features
• Unsupervised many different clustering
  techniques available (including ISODATA)
• Supervised better tools for capturing and
  analyzing signatures, many classification methods
  (including paralleliped and maximum likelihood)
• Harder to use (with power comes complexity)

72
(No Transcript)
73
Critical Elevations for Shrub Establishment

• What are the areas of an island that can be
  colonized by shrubs?
• What level of land elevation must be achieved
  before shrubs colonize?

74
Analysis

• Identify shrub locations
• 1x1 m resolution from 1991 image is sufficient to
  resolve individual colonizing shrubs

75
Elevations

• The NOAA Airborne LIDAR Assessment of Coastal
  Erosion (ALACE) program flew over Hog Island in
  1997
• 15 cm vertical resolution
• 5 m horizontal resolution

76
Analysis

• Compare actual shrub locations with a similar
  number randomly chosen locations
• 150 shrub locations were identified
• Only one shrub per clump identified to avoid
  effects of vegetative growth
• 150 random locations in the vicinity of the
  shrubs were identified

77
Random Locations
Actual Shrub Locations
78
The elevations of shrub and random locations are
used to statistically assess the use of dune and
swale regions by colonizing shrubs
79
Results

• Shrubs were found at a significantly higher
  elevation than randomly selected locations (p ?
  0.001)