1. DATA INTEGRATION AND ANALYSIS
IMAGE CLASSIFICATION
DATA INTEGRATION AND ANALYSIS
Source & Courtesy: Evren Bakılan

2. What is Remote Sensing?
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.
The practice of deriving information about the earth's land and water surfaces using images acquired from an overhead perspective, using electromagnetic radiation in one or more regions of the electromagnetic spectrum, reflected or emitted from the earth's surface.

3. Applications of Remote Sensing
Meteorology (Weather Prediction)
Climatology
Oceanography
Costal Studies
Water Resources
Geology
Archeology
Land cover\land use
Classification and monitoring of urban, agricultural and marine environments from satellite images

4. Principle of Remote Sensing
Interaction between incident radiation and the targets of interest
Recording of Energy by the Sensor (D)
Transmission, Reception, and Processing (E)
Interpretation and Analysis (F)
Application (G)
Energy Source or Illumination (A)
Radiation and the Atmosphere (B)
Interaction with the Target (C)

5. Reflected Light

6. The “PIXEL”

7. Wavelength (Bands)

8. Band Combinations
3,2,1
4,3,2
5,4,3

9. Feature space image Band 4 Band 3
A graphical representation of the pixels by plotting 2 bands vs. each other
For a 6-band Landsat image, there are 15 feature space images
Band 3
Band 4

10. Each color represents a different “cluster” pixels that may correspond to the land cover classes you are interested in

12. Image Classification Why classify? Make sense of a landscape
Place landscape into categories (classes)
Forest, Agriculture, Water, etc
Classification scheme = structure of classes
Depends on needs of users

13. What is a Classified Image
Image has been processed to put each pixel into a category
Result is a vegetation map, land use map, or other map grouping related features
Categories are defined by the intended use of the map
Can be few or many categories, depending on the purpose of the map and available resources

14. Land Cover Classification
Defining the pieces that make up the puzzle

15. Land cover classification steps
Define why you want a classified image, how will it be used?
Decide if you really need a classified image
Define the study area
Select or develop a classification scheme (legend)
Select imagery
Prepare imagery for classification
Collect ancillary data
Choose classification method and classify
Adjust classification and assess accuracy

16. Image Classification

17. Example Uses Provide context Drive models
Landscape planning or assessment
Research projects
Drive models
Global carbon budgets
Meteorology
Biodiversity

18. Application in Agriculture
A - color infrared photograph of big lake
B - classified image of big lake
C - QuickBird image of big lake
D - classified satellite image

19. Basic Strategy: How do you do it?
Use radiometric properties of remote sensor
Different objects have different spectral signatures

20. Basic Strategy: How do you do it?
In an easy world, all “Vegetation” pixels would have exactly the same spectral signature
Then we could just say that any pixel in an image with that signature was vegetation
We’d do the same for soil, etc. and end up with a map of classes

21. Example: Near Mary’s Peak
Image Classification
Example: Near Mary’s Peak
Input data
(Digital)
Output data
(Thematic)
classification process

22. Image Classification black = water yellow = open/field
dark green = dense forest
light green = sparse forest
bronze = mixed urban
red = dense urban

23. Query Formulation Purpose ?
Query “patch” – pertaining to some semantics,
e.g. mountains
Satellite Image
Database
Ranked
Results
Purpose ?
Geography - Find mountainous regions with snow-caps (low-level semantics).
Forestry – Find forests of a certain density, analyze deforestation (mid-level semantics).
Military – Find air-bases in certain regions of the world (high-level semantics).

24. Basic strategy: Dealing with variability
Classification:
Delineate boundaries of classes in n-dimensional space
Assign class names to pixels using those boundaries

25. Information Classes vs. Spectral Classes
Information classes are categorical, such as crop type, forest type, tree species, different geologic units or rock types, etc.
Spectral classes are groups of pixels that are uniform (or near-similar) with respect to their brightness values in the different spectral channels of the data.

26. Classification Strategies
Two basic strategies
Supervised classification
We impose our perceptions on the spectral data
Unsupervised classification
Spectral data imposes constraints on our interpretation

27. Image Classification Classification Supervised Classification
Unsupervised Classification (Clustering)
No extensive prior knowledge required
Unknown, but distinct, spectral classes
Are generated Limited control
over classes and identities
No detailed information
Statistical Techniques
Distribution Free
Gaussian maximum Likelihood classifier
based on probability
distribution models, which
may be parametric
or nonparametric
Euclidean classifier
K-nearest neighbour
Minimum distance
Decision Tree

28. Classification Approaches

29. Supervised Classification

30. Supervised Classification

31. Supervised Classification
Supervised classification requires the analyst to select training areas where he/she knows what is on the ground and then digitize a polygon within that area…
Mean Spectral Signatures
The computer then creates...
Conifer
Known Conifer Area
Digital Image
Water
Known Water Area
Deciduous
Known Deciduous Area

32. Supervised Classification
Information
(Classified Image)
Mean Spectral Signatures
Multispectral Image
Conifer
Deciduous
Water
Unknown
Spectral Signature of Next Pixel to be Classified

33. Areas of Interest Detection of geometric features
Simple line detection filters
Detection of geometric features
e.g. buildings,
high-pass filtering & tresholding
template matching
(temporal) Change Detection
subtraction, ratio, correlation (movement)
comparison of classified images 2 -1 -1 -1 -1 2 -1 2 -1 -1 2 -1 -1 -1 2 2 -1 -1 -1 2 -1 -1 -1 -1 -1 2 -1 2 2 2 -1 2 -1 -1 -1 -1

34. Areas of Interest Image Segmentation detection of homogenous surfaces
by means of tresholding or edge-detection
Region-growing algorithm
High-pass?
Std-filtering?
NDI?

35. Areas of Interest 1) Line-detection filters 2) Averaging filter
3) tresholding 2 -1 -1 -1 -1 2 -1 2 -1 -1 2 -1
1/9
1/9
1/9 -1 -1 2 2 -1 -1
1/9
1/9
1/9 -1 2 -1 -1 -1 -1
1/9
1/9
1/9 -1 2 -1 2 2 2 -1 2 -1 -1 -1 -1

36. Supervised Classification
“Training”

37. Training Areas

38. Supervised Classification
“Segmentation”

39. Supervised Classification
Common Classifiers:
Parallelpiped
Minimum distance to mean
Maximum likelihood

40. Supervised Classification: Statistical Approaches
Minimum distance to mean
Find mean value of pixels of training sets in n-dimensional space
All pixels in image classified according to the class mean to which they are closest

41. Supervised Classification
Parallelepiped Approach
Pros:
Simple
Makes few assumptions about character of the classes

42. Supervised Classification

43. Supervised Classification: Minimum Distance
Pros:
All regions of n-dimensional space are classified
Allows for diagonal boundaries (and hence no overlap of classes)

44. Maximum Likelihood Classifier
Mean Signature 1
Candidate Pixel
Relative Reflectance
Mean Signature 2
It appears that the candidate pixel is closest to Signature 1. However, when we consider the variance around the signatures…
Blue
Green
Red
Near-IR
Mid-IR

45. Maximum Likelihood Classifier
Mean Signature 1
Candidate Pixel
Relative Reflectance
Mean Signature 2
The candidate pixel clearly belongs to the signature 2 group.
Blue
Green
Red
Near-IR
Mid-IR

46. Supervised Classification
In addition to classified image, you can construct a “distance” image
For each pixel, calculate the distance between its position in n-dimensional space and the center of class in which it is placed
Regions poorly represented in the training dataset will likely be relatively far from class center points
May give an indication of how well your training set samples the landscape

47. Supervised Classification
Some advanced techniques
Neural networks
Use flexible, not-necessarily-linear functions to partition spectral space
Contextual classifiers
Incorporate spatial or temporal conditions
Linear regression
Instead of discrete classes, apply proportional values of classes to each pixel; ie. 30% forest + 70% grass

48. Decision Rules in Spectral Feature Space
Maximum Likelihood
(Discriminant Analysis
Parallelpiped
Minimum Distance
to Means

49. Classified Image

50. Unsupervised Classification

51. Unsupervised Classification
In unsupervised classification, the spectral data imposes constraints on our interpretation
How? Rather than defining training sets and carving out pieces of n-dimensional space, we define no classes beforehand and instead use statistical approaches to divide the n-dimensional space into clusters with the best separation
After the fact, we assign class names to those clusters

52. Unsupervised Classification
Clustering

53. Unsupervised Classification
The analyst requests the computer to examine the image and extract a number of spectrally distinct clusters…
Spectrally Distinct Clusters
Cluster 3
Cluster 5
Cluster 1
Cluster 6
Cluster 2
Cluster 4
Digital Image

54. Unsupervised Classification
Saved Clusters
Cluster 3
Cluster 5
Cluster 1
Cluster 6
Cluster 2
Cluster 4
Output Classified Image
Next Pixel to be Classified
Unknown

55. Unsupervised Classification
The result of the unsupervised classification is not yet information until…
The analyst determines the ground cover for each of the clusters…
???
Water
???
Water
???
Conifer
???
Conifer
???
Hardwood
???
Hardwood

56. Unsupervised Classification
It is a simple process to regroup (recode) the clusters into meaningful information classes (the legend).
The result is essentially the same as that of the supervised classification:
Conif.
Hardw.
Water
Land Cover Map
Legend
Water
Conifer
Hardwood
Labels

57. Unsupervised Classification

58. Evaluating Signatures--Signature Ellipses

59. Sources of Errors Acquisition Data Processing Implementation
(Geometric Aspects,Sensor Systems, Platforms, Ground Control, Scene Considerations)
(Geometric Rectification, Radiometric Rectification, Data conversion)
Implementation
Data Analysis
ERROR
(Quantitative Analysis, Classification System, Data Generalization)
Decision Making
Preprocessing the images by appropriate de-noising and enhancement
algorithms have increased the efficiency of the classification.
Data Conversion
Final Product Presentation
Error Assessment
Spatial Error
Thematic Error
(Sampling Error Matrix
Locational Accuracy…)
(Raster to Vector
Vector to Raster)

60. Unsupervised Classification Post classification sorting - ‘labeling’
Cluster 1
Class 1
Cluster 2
Cluster 3
Class 2
Cluster 4
Cluster 5
Class 3
Cluster 6
Cluster 7
Cluster 8

61. Unsupervised Classification
Pros
Takes maximum advantage of spectral variability in an image
Cons
The maximally-separable clusters in spectral space may not match our perception of the important classes on the landscape

62. Unsupervised Classification Results from Clustering - Spectral Classes

63. Input data is a digital data
Data Analysis
Input data is a digital data
Image Rectification and Restoration Geometric Correction Radiometric Correction Noise Removal
Image Enhancement
The objective is to create “new” images from the original image data in order to increase the amount of information that can be visually interpreted from the data.
Image classification – pixelwise classification
Image Classification

64. ISODATA Procedure Arbitrary cluster means are established,
The image is classified using a minimum distance classifier
A new mean for each cluster is calculated
The image is classified again using the new cluster means
Another new mean for each cluster is calculated
The image is classified again...

65. ISODATA Procedure
After each iteration, the algorithm calculates the percentage of pixels that remained in the same cluster between iterations
When this percentage exceeds T (convergence threshold), the program stops or…
If the convergence threshold is never met, the program will continue for M iterations and then stop.

66. ISODATA -- A Special Case of Minimum Distance Clustering
“Iterative Self-Organizing Data Analysis Technique”
Parameters you must enter include:
N - the maximum number of clusters that you want
T - a convergence threshold and
M - the maximum number of iterations to be performed.

67. ISODATA Clusters

68. ISODATA Pros and Cons
Not biased to the top pixels in the image (as sequential clustering can be)
Non-parametric--data does not need to be normally distributed
Very successful at finding the “true” clusters within the data if enough iterations are allowed
Cluster signatures saved from ISODATA are easily incorporated and manipulated along with (supervised) spectral signatures
Slowest (by far) of the clustering procedures.

69. Classification -- Final Thoughts
Classifications are never complete -- they end when time and money run out
Classification is iterative -- it’s tough to get it right the first few iterations
Consider a hybrid classification -- part supervised, part unsupervised
Manual Classification and/or Editing is not cheating!

70. Landsat ETM+ Digital color infrared Acquired: April 21, 2003
Spatial resolution: 30 meters

71. Landsat TM Digital color infrared Acquired: February 17, 1989
Spatial resolution: 30 meters

72. Landsat MSS Digital color infrared Acquired: March 14, 1975
Spatial resolution: 57 meters

73. Corona Panchromatic (b/w) film Acquired: March 2, 1969
Spatial Resolution: 3 meters

74. Examples of Classification Results

75. Scene Classification

76. Some sample patches
Car
Pavement
Road
Tree

77. Thanks…