Satellite Remote Sensing Data And Image

Processing Using IDRISI an Image

processing and GIS Software

M.N.Reddy
PROJECT1 :
Land use/Land Cover Classification from
Remote Sensing Data - IDRISI an Image
processing and GIS Software

PROJECT2 : Characterization:
Characterization of Agro-climatic Zones
in Mahabubnagar District in AP based on
Rainfall and Temperature
IDRISI –not an Acronym

Al-Idrisi–Muslim Scholar
(1100-1166)

Idrisi Project –Ron Eastman of

Clark University, USA, 1987

Clark Labs –1994

 COMPONENTS OF IDRISI

• Cartographic Display System

• Map digitization System (CattaLinx)
• Database management system
(Spatial and Attribute )
• Geographical Analysis System
• Statistical Analysis System
• Image Processing System
• Decision Support System
 Display

Digitization
Decision
support Database

Geographic
Statistical analysis
analysis
Data Products

• Photographic Products
• Digital Products

•A digital image is defined as a

matrix of digital numbers(DNs).
Classification of Data Products

• Raw data

• Partially corrected products

• Geo-coded Products

• Precision Products
 Information extraction
• Visual Interpretation

• Classification
Image Processing:

Act of examining images for the

purpose of identifying objects
and judging their significance
Image Scale and Resolution

Scale : number of unit on ground

units represented by a single unit
on the image

Resolution: Ability of an imaging

system to record fine details in a
distinguishable manner
Example: Suppose you have a IRC 1D
image of 1000 rows and 1000 columns.
If you print this image on a paper of
100 cm by 100 cm in size, what would
be the scale of the printed image?
(Resolution of IRS 1D = 23.5 m)
 PROJECT1 :
Land use/Land Cover Classification from
Remote Sensing Data - IDRISI an Image
processing and GIS Software
Available Data
1. Raw satellite data bearing scene number
TRPC20026J099-059 of part of Medak ditrict in AP

2. Scanned Survey of India Toposheets 56F11, 56F12

in 1:50,000
1. Reference Points of the Images for
rectification(Geo-referencing)
Steps

1. Importing Digital data to IDRISI

2. Image Rectification/Geo-
referencing

3. Rectification of Toposheets (Two)

4. Subset the Image with respect to

Toposheets
5. Registration of subset Image with
respect to Subset Toposheets

6. Mosaic the Images (Joining)

7. Unsupervised/Supervised
Classification

8. Raster to Vector conversion

 Exercise Details

1. Create a Data Folder (working Directory)&

Setting Project Environment

2. Importing Satellite Data –Band by band and

Composite all 4 bands

3. Study Layer Properties – Image properties

5. Creating Imagery Group Files

6. Image Rectification with already available data

in leader file –5 known Lat, lan Points
7. Importing Toposheet 5611.jpg to Idrisi Fomat
8. Rectification / Geo-referencing Toposheet

9. Subsetting the Toposheet 56F11

10. Converting Vector file of Subset toposheet to

Raster
11. Importing Toposheet 56F12.jpg to Idrisi
Fomat
12. Rectification / Geo-referencing Toposheet
56F12
13. Subsetting the Toposheet 56F12

14. Converting Vector file of Subset toposheet to

Raster
15. Overlay of Toposheet 56f11 with rectified
Image9959

16. Overlay of Toposheet 56f12 with rectified

Image9959
17. Composite the files –subset56f11_1 /3

18. Composite the files –subset56f12_1 /3

19. Grouping the files –subset56f11_1 /3 and

subset56f12_1 /3

20. Registration of Image with reference to

Toposheets
21. Rectification of clipped image with respect to
Toposheet

22. Mosaic of Images

23. Classification of images

Geometric Correction:The transformation of a
Remotely Sensed image so that it has the scale
and projection properties of a map is called
geometric correction (Mather, 2002)

• GCP (Ground Control Points)

• Rectification
• Re-sampling
• Registration
GCP: Ground Control Points:
Identification of geographical features
on he image is called GCP

Positions are known as intersection of

streams, highways,airport, runways etc.
Latitudes and Longitudes can be
determined by accurate base maps.
Rectification: Rectification (rubber
sheeting is the process of removing
distortion from imagery by
wrapping the image to fit map
projection

Each pixel is assigned a map coordinate

during rectification
Registration Is the process of making
image data to confirm to another
image
 Resampling method:
The location of output pixels derived from
the ground control points (GCPs) is used
to establish the geometry of the output
image and its relationship to the input
image.
Difference between actual GCP location
and their position in the image are used
to determine the geometric tranformation
 Six-pixel, 3-band Digital Image
56 58 62 BAND1
69 82 94 BAND2
134 135 129 BAND3
14 197 152 BAND1
156 157 143 BAND2
120 172 184 BAND3
 Resolution Types

•Spatial resolution

•Spectral resolution

•Radiometric resolution
•Temporal resolution
Remote Sensing is the science and art of
obtaining information about an object,
area or phenomenon through an analysis
of the data acquired by a device which is
not in contact with the object, area or
phenomenon under investigation.
Spatial resolution

Spatial resolution refers to the fineness of

details visible in an image

It is the spatial resolution of a sensor that

determines the level of spatial details that it
provides about features on the Earth’s surface
Spectral resolution

Spectral resolution refers to the width of the

spectral bands. It is the width across the
electro magnetic spectrum that the

It is the spatial resolution of a sensor that

determines the level of spatial details that it
provides about features on the Earth’s surface
Radiometric resolution

Radiometric resolution refers to the ability of a

remote sensing system to record many levels
of values

It refers to the number of digital values used to

express the data collected by the sensor. It is
commonly expressed as the number of bits
needed to store the maximum level.
Temporal resolution

Temporal resolution is the imaging revisit

interval. It is the frequency with which imges of
a given geographical location can be aquired

The temporal resolution is determined by

orbital characteristics and swath width etc..
Data Processing and Remote sensing
Steps

• Pre-processing

• Display and enhancement

• Information Extraction
 Pre-processing

• Earth rotation correction

• Noise reduction
• Radiometric Correction
• Atmospheric Correction

• Geometric correction
 Image Enhancement
• Contrast Stretching

• Density Slicing
• Colour composites
• Ratio Images
• Principal Components
• Convolution Filtering
• Edge Enhancement and Linear Filtering

• Colour space Transformation

 Image Classification

Is the process of identification of the

patterns associated with each pixel
position in an image in terms of the
characteristics of the objects at the
corresponding point on the Earth’s
surface.
 Land classification
• Aims to label each pixel in a scene
to specific land cover types.
• Pixels can then be either correctly
classified, incorrectly classified or
unclassified.
• Two main type of classification
– Unsupervised
– Supervised
Un-supervised Classification is a process
of grouping pixels that have similar
spectral values
•No previous knowledge assumed about data.
•Tries to spectrally separate the pixels.
•User has controls over:
–Number of classes
–Number of iterations
–Convergence thresholds
•Two main algorithms: Isodata and k-means

Each group of similar pixels is called spectral

class
 Example : spectral plot

• Two bands of data.

• Each pixel marks a
location in this 2d
spectral space
Band 2

• Our eye’s can split

the data into
clusters.
• Some points do not
Band 1 fit clusters.
 Example k-means

Band 2

Band 2
Band 2

Band 1 Band 1 Band 1

2. Second
1. First iteration. 3. N-th iteration.
iteration. The
The cluster centres The centres have
centres move to
are set at random. stabilised.
the mean-centre of
Pixels will be
all pixels in this
assigned to the
cluster.
nearest centre.
 ISODATA (unsupervised)

• Extends k-means. Also calculate standard

deviation for clusters.
• After stage 3 we can either:
– Combine clusters if centres are close.
– Split clusters with large standard
deviation in any dimension.
– Delete clusters that are to small.
• Then reclassify each pixel and repeat.
• Stop on max iterations or convergence
limit.
• Assign class types to spectral clusters.
 Example ISODATA

Band 2
Band 2
Band 2

Band 1 Band 1
Band 1

1. Data is clustered 2.Cyan and green 3. Either assign

but blue cluster is clusters only have 2 outliers to nearest
very stretched in band or less pixels. So cluster, or mark as
1. they will be removed. unclassified.
 Supervised Classification

Is to sample areas of known cover

types to determine representative
spectral values of each cover type.

The samples are referred – Training

fields
Representative spectral values –
Spectral
signatures
 Supervised Classification
Procedures
• Maximum-likelihood Method
• Discriminant Function
• Bayesian Method
• Parallelepiped
• Minimum distance
• Neural network
 Parallelepiped (supervised)
• For each training region determine the
range of values observed in each band.
• These ranges form a spectral box (or
parallelepiped) which is used to classify
this class type.
• Assign new image pixels to the
parallelepiped which it fits into best.
• Pixels outside all boxes can be
unclassified or assigned to the closest
one.
• Problems with classes that exhibit high
correlation between bands. This creates
long ‘diagonal’ data-sets that don’t fit well
into a box.
 Parallelepiped example
Training classes plotted in spectral
space. In this example using 2 bands.
 Maximum likelihood
(supervised)
• For each training class the spectral variance and
covariance is calculated.
• The class can then be statistically modelled with a
mean vector and covariance matrix.
• This assumes the class is normally distributed.
Which is generally okay for natural surfaces.
• Unidentified pixels can then be given a
probability of being in any one class.
• Assign the new pixel to the class with the highest
probability – or unclassified if all probabilities
low.
Concept of Maximum Likelihood Classification
 Maximum likelihood example
• Normal probability
distributions are fitted to
each training class.
• The lines in the diagram
show regions of equal
probability.
1
• Point 1 would be assigned to
class ‘pond culture’ as this
is most probable. Equiprobability
contours
• Point 2 would generally be
unclassified as the
2
probabilities of fitting into
one for the classes would be
below threshold.
Problem:
Some people recommend rectifying an image after
it has been classified. The argument is
3) Rectification is quicker since each pixel
contains only one class value instead of many
spectral values.
4) Some spectral integrity is lost during the pixel
resembling process – an un-rectified image is is
spectrally more correct than a rectified image.

If you do not agree give the reasons to rectify an

image for before classification ,say, vegetation
mapping.