DGI-108:

Practical –III
Digital Image
Processing
 CONTENT

SERIAL PAGE
NAME OF CHAPTER
NO NO

Theory for background 0-42

CHAPTER ONE (SATELLITE

1 3-5
IMAGE DOWNLOAD)
CHAPTER TWO (LAYER
2 6-7
STACK)
CHAPTER THREE (BASIC
3 8-10
FUNCTION)
CHAPTER FOUR
(MULTISPECTRAL BAND
4 11-12
COMBINATION& IMAGE
ENHANCEMENT)
5 CHAPTER FIVE(SUBSET) 13-18

CHAPTER SIX (IMAGE

6 19-24
PREPROCESSING)
CHAPTER SEVEN (IMAGE
7 25-32
INTERPRETATION)
CHAPTER EIGHT (MAP
8 33-40
PRODUCTION

2|Page
I NTRODUCTION TO
1
Digital Image Processing of Remote
Sensing Data
D IGITAL I MAGE P ROCESSING

A digital remotely sensed image is typically composed of picture elements (pixels) located at
the intersection of each row i and column j in each K bands of imagery. Associated with each pixels
a number known as Digital Number (DN) or Brightness value (BV), that depicts the average radiance
of a relatively small area within a scene. A smaller number indicates low average radiance from the
area and the high number is an indicator of high radiant properties of the area. The size of this area
effects the reproduction of details within the scene. As pixel size is reduced more scene detail is
presented in digital representation.
Digital Image Processing is a collection of techniques for the manipulation of digital images by
computers. It started in 1960 when first airborne multispectral scanner data and digitized aerial
photographs became available for analysis. With the launch of Landsat satellite in 1972, the digital
image data became widely available for remote sensing applications. Today access to low cost
efficient computers and satellite data has made this technology very popular for natural resource
management. The core idea in digital image processing is to analyse digital data which is referred as
image by computer systems called as image processing systems. Image processing systems consists
of specialized hardware and software for the analysis and display of digital images.
Digital Image Processing is an extremely broad subject and involves procedures which are
mathematically complex. The procedure for digital image processing may be categorized into the
following types of computer assisted operations.
i. Importing of digital remote sensing satellite data
ii. Pre-processing of Image – Geometric and radiometric corrections
iii. Image enhancement
iv. Image transformation
v. Image classification
vi. Post classification accuracy assessment
There are many digital image processing software available such as ERDAS Imagine, Geomatica,
ENT MIPS, Idrisi, ENVI etc. This manual explains basic digital image processing steps using the
software ENVI. ENVI stands for Environment for Visualizing Images and developed by ITT, Japan
and latest version of ENVI is 4.8. It assumes that you are already familiar with general image-
processing concepts. In order to run this tutorial, you must have ENVI installed on your computer.

Working with ENVI

Start ENVI from the START Menu. Select Start > Programs > ENVI 4.8 >ENVI. the ENVI
Main Menu appears in the form of a button menu bar as given below.
ENVI uses a graphical user interface (GUI) to provide point-and-click access to image processing

1
functions. When you start ENVI, Clicking with the left mouse button on any of the Main Menu
buttons brings up a menu of options, which may in turn contain submenus with further options. The
choices selected from these submenus will often bring up dialog boxes that allow you to enter
information or set parameters relating to the ENVI function you have selected.
You select menu choices and functions using a three-button mouse. However, if you are using
ENVI for Windows with a two-button mouse, you can simulate a “middle” mouse button press by
holding down the Control key and pressing the left mouse button.

ENVI File Formats

ENVI uses a generalized raster data format consisting of a simple flat-binary file and a small
associated ASCII (text) header file. This file format permits ENVI to use nearly any image file,
including those that contain their own embedded header information. ENVI also supports a variety
of data types: byte, integer, unsigned integer, long integer, unsigned long integer, floating-point,
double-precision floating-point, complex, double-precision complex, 64-bit integer, and unsigned
64-bit integer.
Generalized raster data are stored as a binary stream of bytes in either Band Sequential (BSQ),
Band Interleaved by Pixel (BIP), or Band Interleaved by Line (BIL) format.
• BSQ is the simplest format, with each line of data followed immediately by the next line of the
same spectral band. BSQ format is optimal for spatial (x,y) access to any part of a single spectral
band. Many researchers like this format because it is not necessary to read “serially” past
unwanted information if certain bands are of no value.
• In BIP, data for the pixel in all bands are written together. Taking the example of LANDSAT -
MSS (Four Bands of Image Data every element in the matrix has four pixel values (one from
each spectral band) placed one after the other [i.e., pixel (1,1) of band 1, pixel (1,1) of band 2,
pixel (1,1) of band 3, pixel (1,1) of band 4, and then pixel (1,2) of band 1, pixel (1,2) of band 2
and so on]. Again, this is a practical data format if all bands are to be used, otherwise it would be
inefficient. This format is not much popular now, but was used extensively by EROS Data
Centre for Landsat scene at initial stage. BIP format provides optimal spectral processing
performance.
• In BIL format, the data for the bands are written line by line onto the same tape (i.e. line 1 band
1, line 1 band 2, line 1 band 3, line 1 band 4, etc.). It is a useful format if all the bands are to be
used in the analysis. If some bands are not of interest, the format is inefficient since it is necessary
to read serially past all the unwanted data. Format provides a compromise in performance between
spatial and spectral processing and is the recommended file format for most ENVI processing
tasks
ENVI supports a variety of data types: byte, integer, long integer, floating-point, double-precision
floating-point, complex, and double-precision complex.
The separate text header file provides information to ENVI about the dimensions of the image,
any embedded header that may be present, the data format, and other pertinent information. The
header file is normally created (with your input) the first time a particular data file is read by ENVI.
You can view and edit it at a later time using the Edit ENVI Header function under the ENVI File
menu. You can also generate ENVI header files outside ENVI, using a text editor, if desired.

2
ENVI Windows and Displays
As you work with ENVI, a number of different windows and dialog boxes will appear on your
screen. These allow you to manipulate and analyze your image. The most important of these displays
is a group of three windows that display your image, allow you to move around in it, and allow you
to magnify different areas. This group of windows is collectively referred to as the “Display group”.
The Display group consists of:
• The Main Display Window: This is where all or part of your image is displayed at full resolution.
• The Scroll Window: If your entire image does not fit in the Main window, the Scroll window
will appear. The Scroll window displays a reduced-size version of the entire image, which allows
you to select the portion that is displayed in the Main window. A coloured box in the Scroll
window indicates the spatial location and coverage of the Full-Resolution Main Display window.
A number in the title bar of the Scroll Window tells you what reduction factor has been applied
to the image to display the full spatial extent within the Scroll Window.
• The Zoom Window: This window displays an enlarged version of a selected portion of the
Main window. A coloured box in the Main Image Display indicates the spatial location and
coverage of the Zoom window. A number in the title bar of the Zoom Window tells you what
zoom factor has been applied to the image.
You may have any number of displays open on the screen at any time. There are a wide variety
of other types of ENVI windows you may work with, including scatter plots, spectral profiles
interactive stretching, histogram, spectral library etc.

Scroll Window

Main Display Window

Zoom Window

3
Menus in ENVI Windows
The ENVI Main Display Window has its own internal menus,
which provide access to interactive display and analysis functions.
For uncluttered viewing, these menus do not appear automatically;
you must call them up by pressing the left mouse button while the
mouse pointer is inside the window. Once the window menu is on
screen, you can select options from it as you do from any other
ENVI menu. These window-specific menus are dismissed from
the window by pressing the clicking the left mouse button again
outside of the menu.
Different basic ENVI functions will be discussed while covering
various steps of image processing.

4
IMPORTING SATELLITE DATA AND FAMILIARIZATION WITH
BASIC FUNCTIONS OF ENVI
Start ENVI
To open ENVI, select Start > Programs >ENVI 4.8> ENVI.
The ENVI Main Menu appears when the program has successfully loaded and executed.

Open an Image File

To open a satellite image file, one is supposed to know which satellite, sensor and data format.
Select File>Open External File>IRS>Super Structured for opening superstructed format data of
IRS satellite data as given below. You may see also list of many sensors/satellites options to import.

Select image file stored in defined location/folder in the

computer.

Then the file will open in available band list display window

The Available Bands List

ENVI provides access to both image files and to the individual
spectral bands in those files in the available bands list that contains a list
of all the available image bands in all open files.
Use the Available Bands List to load both colour and gray-scale
images into a display by starting a new display or selecting the display
number from the pull down list of displays, clicking on the appropriate
radio button, then selecting the desired bands from the list by clicking
on the band name(s)
The File button on the menu bar at the top of the dialog provides
access to file opening and closing, file information, and the Cancel button.
The Options button provides a function to find the band closest to a
specific wavelength, allows toggling between full and shortened band
names in the list, setting display border sizes, and “folding” all of bands
in a single open image into just the image name. Folding the bands into
single image names can also be accomplished using the Fold button at
the bottom of the dialog.

5
Display of Image
From the available band list, one can choose open a single band as gray scale image or minimum
three bands to develop color image. Display of grayscale image is done as given below

Display Grayscale Image

Display Colour Image

If the Available Bands List is not already on your screen, call it up by selecting Window
>Available Bands List. Load a color image in the second display by clicking on the toggle button
labelled “RGB Colour” in the Available Bands List dialog. Select a band for each colour from the

6
list and click “Load RGB” to load the image as shown below. While displaying the different bands
of a multispectral data set, images obtained in different bands are displayed in image planes (other
than their own) the colour composite is regarded as False Colour Composite (FCC). For a true
colour composite an image data used in red, green and blue spectral region must be assigned bits of
red, green and blue image processor frame buffer memory. A colour infrared composite ‘standard
false colour composite’ is displayed by placing the infrared, red, green in the red, green and blue
frame buffer memory.
In this healthy vegetation shows up in shades of red because vegetation absorbs most of green
and red energy but reflects approximately half of incident Infrared energy. Urban areas reflect equal
portions of NIR, R & G, and therefore they appear as steel grey.

7
Saving the displayed Image into ENVI Standard Format
Displayed image will be saved as ENVI standard image format for its display, use later. The
steps followed is given as below.

The image is now saved in the ENVI Standard format and can now be directly opened by:
File>Open Image File>Select your Image File from the location where the imported file was saved
in STEP 5>OK

8
Alternative way for opening the image if the data type is not known
If the image with you is not compatible with the listed data format of satellite sensors given or
not known how to open, we can import it provided we have its minimum information required such
no of samples (pixel), lines (row), band, Offset (header size), interleave (data format) as given in
below in header info window. Following steps are followed to retrieve these information and display
the image.
Step to know minimum head information required for display

9
Steps for retrieving the information required

For example retrieval of information required for Image Analysis of IRS Standard product
(Note: information from imagery file is indicated by ( ) and from leader file by [ ])
(a) No. of rows : (i.e. no of lines)
1. No. of lines per image : ( 237-244) or [ 7417-7432]
(Excluding top and bottom border lines)
2. No. of top border lines : (261-264)
3. No. of bottom border lines : (265-268)
4. Total no. of lines per image (no. of rows) : [1 +2 +3]
Alternative
1. No. of image records : (181-186)
2. No. of bands : (233-236)
3. No. of records per band (i.e. no of rows) : [1/2]

10
(b) No. of Columns (pixels)
Image record length : (187-192)
Alternatives
1. No. of image pixels per line : (249-256) or [7401-7416]
2. No. of right border pixels per line : (257-260)
3. No. of left border pixels per line : (245-248)
4. No. of bytes of suffix data per record : (289-292)
5. No. of bytes of prefix data per record : (277-280)
6. Total no. of pixels : [1+2+3+4+5]
(c) Header file size Image file size – (Pixels x rows x bands)
(d) Geographic Location (Source file : Leader.*)
Top left (1st corner) Top right (2nd corner)
Latitude [6269-6284] [6317-6332]
Longitude [6285-6300] [6333-6348]
Line no. [6301-6308] [6349-6356]
Pixel no. [6309-6316] [6357-6364]
Bottom left (3rd corner) Bottom right (4th corner)
Latitude [6365-6380] [6413-6428]
Longitude [6381-6396] [6429-6444]
Line no. [6397-6404] [6445-6452]
Pixel no. [6405-6412] [6453-6460]
Scene centre location
Latitude [6221-6236] Line no. [6253-6260]
Longitude [6237-6252] Pixel no. [6261-6268]
Map projection type : [24501-24506]
Name of reference ellipsoid [24507-24522]
(e) Interleaving indicator (format type) (269-272) or [7449-7464]
(f) Other information
Name of Satellite [6950-6965]
Time of Satellite pass [6157-6188]
Sensor identification [6966-6997]
Path no. [6141-6148]
Row no. [6149-6156]

11
 Spectral mode [6998-7013]
Sun angle (in degrees)
Azimuth [6694-6709]
Elevation [6710-6725]
No. of bands [7233-7240]
Wavelength range (nm)
Band2 [7241-7272]
Band3 ‘’
Band4 ‘’
Band5 ‘’
Spectral band indicator [7465-7480]
No. of bits per pixel (217-220)
Maximum data range (441-448)
No. of pixels per data group (221-224)
No. of bytes per data group (225-228)
Processed pixel spacing (meters) [7433-7440]
Processed line spacing (meters) [7441-7448]
No. of line losses [7753-7768]
No. of dead detectors [7769-7784]
Satellite altitude (meters) [16253-16268]
Yaw value (in degrees) [16341-16350]
Roll value [16741-16750]
Pitch value [17141-17150]
On board gain no. [7206-7224]
Minimum (Lmin) and Maximum (Lmax) radiance [7337-7400]
(mw/sq.cm/str/micrometer) for no. bands present
in the product
Conversion formula for DN to radiance:
Lrad = (DN/max gray) * (Lmax - Lmin )+ Lmin
Where,
Lrad : radiance for a given DN Value,
DN : Digital Count,
Max gray : 63 for PAN, 127 for WiFS and LISS –3 for raw products only, and 255 for corrected
products
Lmin/ Lmax : minimum/maximum radiance value for a given band.

12
Familiarize with other options available

Display the Cursor Location

• To display the cursor location and value, select Basic Tools -> Cursor Location / Value from the
ENVI Main Menu.
A dialog box will appear displaying the location of the cursor in the Main, Scroll, or Zoom
windows The dialog also displays the screen value (color) and the actual data value of the pixel
underneath the cross-hair cursor.
• To dismiss the dialog, select “Cancel” from the pulldown File menu at the top of the dialog.

Display Image Profiles

X (horizontal), Y (vertical), and Z (spectral) profile plots can be selected and displayed
interactively. These profiles show the data values across an image line (X), column (Y), or spectral
bands (Z).
1. Click the right mouse button anywhere in the Main window to display the Functions menu
button and press and hold the left mouse button on the Functions button to pull down the
menu.
2. Select Functions-> Profiles-> X Profile. to display a window plotting data values versus sample
number for a selected line in the image Repeat the process, selecting Y Profile to display a plot
of data value versus line number, and selecting Z Profile to display a spectral plot Position the
plot windows so you can see all three at once.
A red cross-hair extends to the top and bottom and to the sides of the Main window. The red
lines indicate the line or sample locations for the vertical or horizontal profiles.
1. Move the cross-hair around the image (just as you move the zoom indicator box), to see how
the three image profile plots are updated to display data on the new location. The Mouse Button
Descriptions dialog can be used to guide your exploration of these windows.
2. Close the profile plots by selecting File->Cancel from within each plot window.

Perform Interactive Contrast Stretching

Interactive Contrast Stretching plots a histogram and allows you to interactively control the
contrast of the displayed image. Many different types of stretches can be applied. By default, a
linear 2% stretch is applied to the data when it is first displayed.
• To access ENVI’s interactive contrast stretching functions, select Functions -> Display
Enhancements-> Interactive Stretching.
A dialog that allows you to change the contrast stretch of the displayed image appears Two
histogram plots display the color or grayscale range of the input image (left) and the output image
after contrast stretching (right). Initially, the input and output histograms reflect the default stretch
applied to the data when the image was displayed.
• The Stretch_Type pulldown menu at the top of the histogram has a variety of contrast-stretching
options. Try applying the following methods described below and observe the results in the
Main window.
• Also, try selecting both Zoom and Scroll from the Histogram_Source pulldown menu and note
the differences in the histograms and stretches.

13
Linear
When images are loaded into the Main window, a 2% linear contrast stretch is applied by default.
• Select Functions-> Display Enhancements-> Interactive Contrast Stretching in the main display, then
choose Linear from the pulldown Stretch_Type menu item in the dialog
Note the red and green vertical lines in the input histogram plot—these bars can be repositioned
to control the minimum and maximum value used in the contrast stretch.
• Position the mouse cursor on the red bar and hold down the left mouse button, then drag the bar
from side to side.
Numbers above the plot display the current data value, the number of pixels and the percentage
of pixels that have that value, and the cumulative percentage of pixels with values less than or equal
to the current value.
1. Try positioning the red bar so that approximately 5% of the pixels are selected, and the green
bar so that approximately 95% of the pixels are selected.
2. If Options->Auto Apply Off is selected (the default), the stretch will automatically be applied.
Otherwise, click “Apply” to apply the stretch shown in the output histogram plot to the image
in the Main window.
3. You can also position the bars by entering minimum and maximum values in the fields next to
the “Apply” button at the top of the dialog. You can enter either data values or percentages.
4. Enter “4%” in the left box and “96%” in the right box (press Enter after typing each value) and
click “Apply”.
The red and green bars will be adjusted and the two text entries updated to display the data
values at 4% and 96%, respectively.

Equalize
1. Select Stretch_Type->Equalization and note the change in the “Output Histogram” display.
2. If Options->Auto Apply Off is selected (the default), the stretch will automatically be applied.
Otherwise, click “Apply” to apply the stretch shown in the output histogram plot to the image
in the Main window.

Gaussian
1. Select Stretch_Type -> Gaussian.
2. Set the Standard Deviation by selecting Options -> Set Gaussian Stdv.
3. If Options->Auto Apply Off is selected (the default), the stretch will automatically be applied.
Otherwise, click “Apply” to apply the stretch shown in the output histogram plot to the image
in the Main window.
4. Click “Cancel” to close the contrast stretching dialog.

Display Interactive Scatter Plots

You can plot the data values of two selected image bands versus each other in a scatter plot to
graphically display the overlapping values.

14
1. Select Functions->Interactive Analysis->2D Scatter Plots in the Main window.
A dialog will appear that lets you choose two image bands to compare.
2. Select one band for the X axis and another band for the Y axis and click “OK.”
It may take a few seconds for ENVI to extract and tabulate the data values.
• Once the scatter plot has appeared position the mouse cursor anywhere in the Main window and
drag with the left mouse button pressed.
Pixel values contained in a ten-pixel by ten-pixel box surrounding the crosshair will be highlighted
in red on the scatter plot. Again, note that the Mouse Button Descriptions dialog can help you know
the functions of the different mouse buttons in the Scatterplot.
• Move the cursor around in the Main window to observe the “dancing pixels” effect.
You can also use the scatter plot to highlight specific data values in the Main window.
• Place the mouse cursor in the scatter plot window and click and drag with the middle mouse
button.
A ten-pixel-square box will appear in red on the plot. Pixels with the values contained in the box
are highlighted on the image in the Main window as “Image Dancing Pixels”.
• Select File -> Cancel to close the scatter plot window.

Apply a Display Filter

ENVI gives you the ability to apply several different pre-defined or user-defined filters to a
display (file-based filtering is also available and is accessed via the Filter menu on the ENVI main
menu). The following example shows you how to apply a pre-defined filter to the image in the Main
display window.

Choose a Filter
1. Select Functions->Display Enhancements->Filter and choose the desired filter type from the pulldown
filter menu to apply a filter to the displayed image.
2. Try the different sharpening, smoothing, median, and edge detection filters on the displayed
image.

Display A Second Image and Apply Filter

• Select “New Display” from the pulldown Display menu button at the bottom of the Available
Bands List dialog to create a second display group, select an image band, and click “Load Band”
to load the image into the second display. Choose Functions->Display Enhancements->Filter and
select a filter different from that applied to image #1 from the pulldown filter menu.

Link Two Displays

Link the two displays together to compare the filter results. When you link two displays, any
action you perform on one display (scrolling, zooming, etc.) is echoed in the linked display. To link
the two displays you have on screen now:
1. Select Functions ->Link- >Link Displays from one of the display windows.

15
2. Click “OK” in the Link Displays dialog.
3. Now try scrolling or zooming in one display group and observe as your changes are mirrored in
the second display.

Dynamic Overlays
ENVI’s Multiple Dynamic Overlay feature allows you to dynamically superimpose parts of one
or more linked images onto the other image. Dynamic overlays are turned on automatically when
you link two displays.
• To use the overlay, position the mouse cursor anywhere in either Main window and hold down
and drag with the left mouse button.
A small portion of the linked image will be superimposed on the current image window. The
overlay will appear in either the Main window or the Zoom window.
• To change the size of the overlay area, hold down the middle mouse button and drag the overlay
area until it is the desired size.

Load a Color Image

1. If the Available Bands List is not already on your screen, call it up by selecting File-> Available
Bands List
2. Load a color image in the second display by clicking on the toggle button labelled “RGB Color”
in the Available Bands List dialog.
3. Select a band for each color from the list and click “Load RGB” to load the image.
4. Try the dynamic overlay feature with the color and grayscale images.
5. Dismiss the first display by clicking “Cancel”.

Select Regions of Interest

ENVI lets you define “Regions of Interest” (ROIs) in your images. ROIs are typically used to
extract statistics for classification, masking, and other operations.
1. Select Basic Tools->Region of Interest->Define Region of Interest from the ENVI Main Menu. The
ROI Definition dialog will appear
2. Draw a polygon that represents the region of interest.
3. Click the left mouse button in the Main window to establish the first point of the ROI polygon.
Select further border points in sequence by clicking the left button again, and close the polygon
by clicking the right mouse button.The middle mouse button deletes the most recent point, or (if
you have closed the polygon) the entire polygon.
• ROIs can also be defined in the Zoom and Scroll windows.
When you have finished defining an ROI, it is shown in the dialog’s list of Available Regions,
with the name, region color, and number of pixels enclosed
1. To define a new ROI, click “New Region”.
2. You can enter a name for the region and select the color and fill patterns for the region by
clicking on the “Edit” button.

16
Working with ROIs
You can define as many ROIs as you wish in any image
• Once you have created the ROI definitions, you can erase them from the display (leaving the
definition in the list) by selecting the ROI from the list and clicking on the “Erase” button.
• Clicking on the “Stats” button allows you to view statistics about the ROI you select.
• Clicking the “Delete” button permanently deletes ROI definitions from the list.
• The other buttons and options under the pulldown menus at the top of the ROI Definitions
dialog let you calculate ROI means, save your ROI definitions, load saved definitions, or display
or delete all the definitions in the list.
Region Of Interest definitions are retained in memory after the ROI Definition dialog is closed,
unless you explicitly delete them. This means the ROIs are available to other ENVI functions even
if they are not displayed.

Save and Output an Image

ENVI gives you several options for saving and outputting your filtered, annotated, gridded images.
You can save your work in ENVI’s image file format (Figure 20), or in several popular graphics
formats (including Postscript) for printing or importing into other software packages. You can also
output directly to a printer.

Saving your Image in ENVI Image Format

To save your work in ENVI’s native format (as an RGB file):
• Select Functions->Output Display ->Image File in the Main window.
The Output Display to Image Parameters dialog will appear.
• Select 24-Bit color or 8-Bit grayscale output, graphics options (including annotation and gridlines),
and borders.
If you have left your annotated and gridded color image on the display, both the annotation and
grid lines will be automatically listed in the graphics options.
You can also select other annotation files to be “burned in” to the output image.
1. Select output to “Memory” or “File”.
2. If output to “File” is selected, enter an output filename.
3. Click “OK” to save the image.
4. If you select other graphics file formats from the “Output File Type” button, your choices will
be slightly different.

End the ENVI Session

When you have finished your ENVI session, click “Quit” or “Exit” on the ENVI Main Menu.

17
 RECTIFICATION AND IMAGE GEO REFERENCING
Rectification is a process of geometrically correcting an image so that it can be represented on
a planar surface, conform to other images or conform to a map. That is, it is the process by which
geometry of an image is made planimetric. It is necessary when accurate area, distance and direction
measurements are required to be made from the imagery. It is achieved by transforming the data
from one grid system into another grid system using a geometric transformation.
Rectification is not necessary if there is no distortion in the image. For example, if an image file
is produced by scanning or digitizing a paper map that is in the desired projection system, then that
image is already planar and does not require rectification unless there is some skew or rotation of
the image. Scanning and digitizing produce images that are planar, but do not contain
any map coordinate information. These images need only to be geo-referenced, which is a much
simpler process than rectification. In many cases, the image header can simply be updated with new
map coordinate information. This involves redefining the map coordinate of the upper left corner
of the image and the cell size (the area represented by each pixel).
The procedure involves selection of ground control points, selection of transformation model
and resampling technique and georeferencing the data means to map a remote sensing image to a
known location on the earth, by referencing it to a map projection.
ENVI supports georeferenced images in numerous predefined map projections including UTM
and State Plane. It also allows to customize map projections utilizing 6 basic projection types, over
35 different ellipsoids, and more than 100 datums to suit most map requirements. ENVI map
projection parameters are stored in an ASCII text file map_proj.txt that you can edit or modify
using ENVI map projection utilities. The information in this file is used in the ENVI header files
associated with each image and allows simple association of a reference pixel location with known
map projection coordinates.
ENVI has following options of image registration for raster images such as image to map and
image to image registration. Following steps are followed.
Image to Map registration: This step is followed when image is to be georeferenced with collected
ground control points through you through key boards or connected GPS.
Step 1: Open the image to be georeferenced following the steps given in Exercise 1
Step 2: Select ENVI>Map>Select GCPs: Image to Map as given below

18
Step 3: Select registration projection window will appear as given here. Select the desired projection
and Click OK, Ground control points selection dialog box appears.

Step 4: Select the known points in the image (will be reflected automatically as Image X and Image
Y in the above window) by moving curser very carefully in zoom window and enter the corresponding
co-ordinates (longitude and latitude) in the black spaces (two on the left side of window) which is
known to you from reference maps, toposheets or GPS collected points. All these points are called

Ground Control Points (GCPs). Then click add points and show list as given here in the window.
Follow the same procedure for all the required points (minimum required is four). Once you have at
least four points, the RMS error is reported. You may repeat collection GCPs if not satisfied and
high RMS value. The GCP collected can be saved clicking File > save table to ASCII.
Step 5: Go to Ground control point selection window >Options>Warp displayed band (for displayed
image only) or Wrap File for whole image having all the bands. Select the image to be georefenced
and click ok to open registration parameters window.
Step 6: Provide required registration (wrap) parameters and give output file name and folder to be
saved in. For example: Method : Polynomial, Degree :1, Resampling Nearest Neighbor

19
Image to Image Registration
This step is followed when one image needs to be georeferenced (wrap image) with another
image which is georeferenced one (base image). First of all, you have to display both the images.
For example, base image is in display 1 and wrap image is in display 2.
Step 1 : Go to Map > Registration > Select GCPs: Image to Image.
The Image to image registration window appears.
Step 2: In the Base Image field, select Display #1. In the Warp Image
field, select Display #2 (the image to be corrected).
Click OK to start the registration. The Ground Control Points
Selection window appears.
Step 3: Individual ground control points (GCPs) can be added by
positioning the cursor in the two images to the same ground location.
Step 4: Identify the same location point on both the display. Click
that point in the base image then click the corresponding point in the
warp image (image file).
Step 5: Examine the locations in the two Zoom windows and
adjust the locations if necessary by clicking the left mouse button
in each Zoom window at the desired locations. Note that sub-
pixel positioning is supported in the Zoom windows. The larger
the zoom factor, the finer the positioning capabilities.
Step 6 : In the Ground Control Points Selection window, click
Add Point to add the GCP to the list. Click Show List to view the
GCP list. Try positioning the cursor at a new location in the Image
window containing the base image. Click the Predict button(It become active only after collecting 3
GCPs) to move the cursor position in the image file to match its predicted location based on the
warp model. The exact position can now be interactively refined by moving the pixel location
slightly in the image file. Click Add Point in the Ground Control Points Selection dialog to add the
new GCP to the list. Once at least 4 points has been selected, the RMS error is reported. Follow the
same procedure for selecting more GCPs. In case wrong GCPs has been selected it can be deleted
by selecting that GCP row in the Image to Image GCP list and Click delete.

20
Step 7 : Once the GCP collection procedure is over it can be saved in ASCII format. Go to Ground
Control Points Selection dialog menu bar, select File > Save GCPs to ASCII (.pts) OR Goto Image
to Image GCP list select File > Save table to ASCII (.txt). To clear all of the points go to Ground
Control Points Selection dialog menu bar, select Options > Clear All Points. From the Ground
Control Points Selection dialog menu bar, select File > Restore GCPs from ASCII.
The following image provides descriptions of the features and functions available in the Image
to Image GCP List. Click Show List to view the GCP list. Click on individual GCPs in the Image to
Image GCP List dialog and examine the locations of the points in the two images, the actual and
predicted coordinates, and the RMS error. Resize the dialog to observe the total RMS Error listed in
the Ground Control Points Selection dialog.

Step 8: Warping Images

The image can be warped from the displayed band, or multiband images can be warped at once.
• From the Ground Control Points Selection dialog menu bar, select Options > Warp Displayed
Band (or Warp File). The Registration Parameters dialog appears.
• Click the Method drop-down list and select RST.
• Ensure the Resampling drop-down list has the Nearest Neighbor option selected.
• In the Enter Output Filename field give new file name and click OK. The warped image is the
final GEOREFERENCED image which gets listed in the Available Bands List when the warp is
completed

21
 The image can also be warped by selecting RST method and Cubic Convolution resampling,
Polynomial method and Cubic Convolution resampling, Triangulation method and Cubic
Convolution resampling, RST method and Bilinear resampling.

Viewing/Editing Map Info in the ENVI Header

Step 1: Right click the map info option of the image file in Available band list window and select
edit map info option. Edit map Information window will open.
This dialog lists the basic map information used by ENVI in
georeferencing. The image coordinates correspond to a reference
pixel used by ENVI as the starting point for the map coordinate
system. Because ENVI knows the map projection, pixel size, and
map projection parameters based on this header information and
the map projection text file, it can calculate the geographic
coordinates of any pixel in the image. You can enter coordinates in
either map coordinates or geographic (latitude/longitude)
coordinates.
Step 2: Click on the projection field toggle button to display the
latitude/longitude coordinates. Click the DDEG button then click
the DMS button to toggle between degrees-minutes-seconds, and
decimal degrees, respectively.
Step 3: Click on Change Proj. and select from listed projections.
Step 4: You can choose to display the location of your mouse cursor,
projection, screen value (RGB color), and the data value etc. of the
pixel underneath the crosshair cursor using the Cursor Location/Value dialog.

22
S UBSETTING AND MASKING TO GET REQUIRED IMAGE
FROM WHOLE IMAGE

Many ENVI tools allow to subset the data before processing. We can perform spatial, spectral,
or statistics subsetting, and, masking of whole image to get image of area of interest. Subsetting is
done cut image of regular shape having maximum extent of the study area where as masking is
done to cut exactly the study area as it looks like.

Spatial Subsetting
You can select spatial subsets by using the following methods:
• Entering samples and line values
• Selecting interactively from the image
• Entering map coordinates
• Using the same spatial subset that was previously used on another file
• Using the image shown in the meta scroll window
• Using the bounding box around a region of interest
Step 1: ENVI> Basic tools > Resize Data. Resize Data Input file opens

Step 2: Select Spatial subset. The options in the Spatial Subset dialog vary depending on whether the
current data is sample-line-based or georeferenced. Additionally, if the same image is open in more
than one display group, you can specify which display number to apply the subset to.

23
Step 3: Subsetting by Samples/Lines

Step 4: Subsetting by Images

To change the subset size or location, select from the following options:
• In the Subset by Image dialog, click and grab on one of the corners of the box and drag to
the desired location.
• To move the box around the image, click on the box and drag it to the desired location.
• Change the values in the Samples or Lines fields.
• Click OK. The starting and ending sample and line coordinates appear in the text boxes
labeled Samples and Lines.

24
Step 5: Subsetting by Map Coordinates
For georeferenced images, you may select the subset by map coordinates or latitude/longitude
coordinates:

To change the map projection of the coordinates, click the corresponding Change Proj. button
and follow the instructions in Selecting Map Projection Types.
Step 6: Subsetting Using Another File’s Subset
There are two options for subsetting a file based on the subset area of a previously subset image:
subset image based on a subset of an image that was originally the same size as the one you are
subsetting and subset image based on an image of the same or smaller size that was previously
subset by map or pixel values.

25
Using the Subset of an Image of the Same Size

Using the Subset of an Image of the Same or Smaller Size

Step 7: Subsetting Using ROIs

To make previously saved ROIs available for selection in the Subset Image by ROI/EVF Extent
dialog, click Open in the Select Spatial Subset dialog, select ROI File or EVF (ENVI Vector File
which is
conversed
from. shp or
other vector
data file) File,
and choose
the desired file
in the
subsequent
Input File
dialog.

26
 To subset a file using the bounding box (i.e., the area encompassing an ROI or group of ROIs)
follow these steps:

27
Step 8: Creating ROIs
Display the image from which ROI is to be created

The following ROI types are available in ENVI: Polygon, Polyline, Point, Rectangle and square,
Ellipse and circle, Multi Part (donut)

28
Step 8: Subsetting by Scroll Window
To subset a file based on the image currently shown in the meta zoomed scroll window for the
selected file, click Scroll in the Select Spatial Subset dialog. The starting and ending values are set to
include only the Scroll window subset.
Subsetting by Scroll window is only available if the Scroll window displaying the selected file is
in meta zoom mode. If the file is displayed in multiple meta zoomed Scroll windows, the subset of
the first window is used.
Spectral Subsetting : It is done to subset the spectral bands as per user requirement from multiband
data. Steps followed are given here with.

29
Masking of the image

Step 1: Building the Mask

• Select Basic tools > Build mask, Input display window appears
• Select display #1 (the display of image to be masked) from select input display window,
mask definition window appears
• Selection options> select desired option and then give output file name and click apply,
• Image having mask will be listed in band list window which is further used for creating mask
image.

Step 2: Creating mask image

• Select Basic tools > Apply mask, Apply Mask Input file window appears
• Select the input file (the image to be masked)
• Select mask band option > select Mask Band from the Select input band window>OK
• Give the new mask image file name and folder to be saved to
• Display the masked image created
• If blank image like display appears without the mask image, then go the image display
window> Enhance> [Image] Linear 2%. You will be able see the image.

Alternate Step
In the following step, mask image is also creating with doing building mask operations using
steps as given below.

This creates Mask file, now Click OK in the Apply Mask input file window. We get desired
subset image through mask. Image may be displayed and subsetting to may be done using ROI or
EVF file used to make the image smaller having only mask image area.

30
I MAGE C LASSIFICATION AND P OST C LASSIFICATION
PROCESSING
Image classification technique assigns the picture elements into various categories. It utilizes
the pixel by pixel spectral information for automated land-cover classification. It makes use of
multiple images of the same scene, obtained over various spectral regions as input and produces
output.
Image classification is based on statistical decision theory. The decision to classify a pixel into
various classes depends upon statistical calculation. It extracts the information from the data. Patterns
i.e. set of radiance measurements obtained in various wavelength bands for each pixel are analysed
and image classification is done. In spatial pattern recognition categorization on image pixels is
done based on their spectral properties, whereas in spatial pattern recognition categorization is done
based on pixels relationship with neighbouring pixels. In temporal pattern recognition multidate
data is analysed to classify a particular category. Classification can be divided in many ways. Some
of them are –

Classification Methods
• Unsupervised Classification Methods-
♣ IsoData,
♣ K-Mean.
• Supervised Classification Methods-
♣ Parallelepiped,
♣ Minimum Distance,
♣ Mahalanobis Distance,
♣ Maximum Likelihood,
♣ Spectral Angle Mapper,
♣ Binary Encoding,
♣ Neural Network,
♣ Support Vector Machine.
• Decision Tree Classification Method

Unsupervised Classification Methods

Unsupervised classification can be used to cluster pixels in a dataset based on statistics only,
without any user-defined training classes. The available unsupervised classification techniques are
K-Means and ISODATA. It is an automated technique that searches for natural groups, or clusters,
of pixels based on their brightness in several bands. Unlike the supervised method, unsupervised
classification does not begin with a predefined set of classes.
• K-Means: A technique that calculates initial class means evenly distributed in the data space,
then iteratively clusters the pixels into the nearest class using a minimum distance technique.
• IsoData: A technique that calculates class means evenly distributed in the data space, then
iteratively clusters the remaining pixels using minimum distance techniques.

31
Applying K-Means Classification
K-Means unsupervised classification calculates initial class means evenly distributed in the data
space, then iteratively clusters the pixels into the nearest class using a minimum-distance technique.
Each iteration recalculates class means and reclassifies pixels with respect to the new means. All
pixels are classified to the nearest class unless a standard deviation or distance threshold is specified,
in which case some pixels may be unclassified if they do not meet the selected criteria. This process
continues until the number of pixel each class changes by less than the selected pixel change threshold
or the maximum number of iterations is reached.

Steps for K-Means Classification

IsoData Classification
ISODATA unsupervised classification calculates class means evenly distributed in the data space
then iteratively clusters the remaining pixels using minimum distance techniques. Each iteration
recalculates means and reclassifies pixels with respect to the new means. This process continues
until the number of pixels in each class changes by less than the selected pixel change threshold or
the maximum number of iterations is reached.

32
Steps for IsoData Classification

Supervised Classification Methods

Supervised classification can be used to cluster pixels in a dataset into classes corresponding to
user-defined training classes. This classification type requires that you select training areas for use as
the basis for classification. Various comparison methods are then used to determine if a specific
pixel qualifies as a class member.
ENVI has broad range of supervised classification techniques, including-
♣ Parallelepiped,
♣ Minimum Distance,
♣ Mahalanobis Distance,
♣ Maximum Likelihood,
♣ Spectral Angle Mapper,
♣ Binary Encoding,
♣ Neural Network,
♣ Support Vector Machine.

33
Selecting Training Sets Using Regions of Interest (ROI)
Display the image from which ROI is to be created

The following ROI types are available in ENVI:

Polygon, Polyline, Point, Rectangle and square, Ellipse and circle, Multi Part (donut)

34
Parallelepiped Classification
Parallelepiped classification uses a simple decision rule to classify multispectral data. The decision
boundaries form an n-dimensional parallelepiped classification in the image data space. The
dimensions of the parallelepiped classification are defined based upon a standard deviation threshold
from the mean of each selected class. If a pixel value lies above the low threshold and below the
high threshold for all n bands being classified, it is assigned to that class. If the pixel value falls in
multiple classes, ENVI assigns the pixel to the last class matched. Areas that do not fall within any
of the parallelepiped classifications are designated as unclassified.

Steps for Parallelepiped Classification

35
Minimum Distance Classification
The minimum distance classification uses the mean vectors of each ROI and calculates the
Euclidean distance from each unknown pixel to the mean vector for each class. All pixels are
classified to the closest ROI class unless the user specifies standard deviation or distance thresholds,
in which case some pixels may be unclassified if they do not meet the selected criteria.

Steps for Minimum Distance Classification

Mahalanobis Distance Classification

The Mahalanobis Distance classification is a direction sensitive distance classifier that uses
statistics for each class. It is similar to the Maximum Likelihood classification but assumes all class
covariances are equal and therefore is a faster method. All pixels are classified to the closest ROI
class unless you specify a distance threshold, in which case some pixels may be unclassified if they
do not meet the threshold.

36
Steps for Mahalanobis Distance Classification

Maximum Likelihood Classification

Maximum likelihood classification assumes that the statistics for each class in each band are
normally distributed and calculates the probability that a given pixel belongs to a specific class.
Unless a probability threshold is selected, all pixels are classified. Each pixel is assigned to the class
that has the highest probability (i.e., the maximum likelihood).

37
Steps for Maximum Likelihood Classification

38
Post Classification Processing
Classified images require post-processing to evaluate classification accuracy and to generalize
classes for export to image-maps and vector GIS. Post Classification can be used to calculate class
statistics and confusion matrices

Extracting Class Statistics

This function allows to extract statistics from the image used to produce the classification.
Separate statistics consisting of basic statistics, histograms, and average spectra are calculated for
each class selected.
1. From the ENVI main menu bar, select Classification > Post Classification > Class Statistics.
The Classification Input File dialog appears.
2. Locate the Classified image and click OK
3. The Statistics Input File dialog box appears.
4. Select the image file and click OK. The Class Selection dialog appears.
5. Click the Select All Items button and click OK. The Compute Statistics Parameters dialog
appears.
6. Click the Basic Stats, Histograms, Covariance, and Covariance Image check boxes in the
Compute Statistics Parameters dialog to calculate all the possible statistics.
7. Click OK to compute the statistics. The Class Statistics Results dialog appears.

39
Confusion Matrix
ENVI’s confusion matrix function allows comparison of two classified images (the classification
and the “truth” image), or a classified image and ROIs. The truth image can be another classified
image, or an image created from actual ground truth measurements.
1. Select the can_pcls.img file and click OK. The Ground Truth Input File appears.
2. Click the Open drop-down button and select New File.
3. Navigate to the envidata\can_tm directory, select the file can_sam.img from the list, and
click Open.
4. Select the can_sam.img file in the Ground Truth Input File dialog and click OK. The Match
Classes Parameters dialog appears.
5. Select Region #1 from both fields and click Add Combination. Continue to pair corresponding
classes from the two images in this way, then click OK. The Confusion Matrix Parameters
dialog appears.
6. Click the Output Result to Memory radio button then click OK.
7. Examine the confusion matrix and confusion images (in the Available Bands List). Determine
sources of error by comparing the classified image to the original reflectance image using
dynamic overlays, spectral profiles, and Cursor Location/Value.

40
Editing Class Colours
When a classification image is displayed, you can change the colour associated with a specific
class by editing the class colours.

41
1. From the Display group menu bar, select Tools > Colour Mapping > Class Colour Mapping.
The Class Colour Mapping dialog appears.
2. Click on one of the class names in the Class Colour Mapping dialog and change the colour by
dragging the appropriate colour sliders or entering the desired data values. Changes are applied
to the classified image immediately.
3. To make the changes permanent, select Options > Save Changes from the menu bar in this the
dialog.

42
 CHAPTER ONE (SATELLITE IMAGE DOWNLOAD)
For downloading satellite image go to https://earthexplorer.usgs.gov. If you don’t have an account then
sign up for opening an account with giving the actual information.

 Search criteria>coordinate> Add coordinate in two format> degree/minute/second or decimal

3|Page
 Or you can also use map by cursor,zoom in to the desire area and add place mark by clicking mouse.
 After adding coordinate go to data range>>select the date range that you want.

 Then select dataset>Landsat>select which Landsat image you want, check the option>result

4|Page
 After showing result you search the date of image that you want and also see various information
about image such as footprint, raw data, band information, preview on the exact location. etc.

 After that there is a download option in the tray and download the TIFF format image for getting all
band.

5|Page
 CHAPTER TWO (LAYER STACK)
 Here the TIFF format image. (LT05_L1TP_136045_20091206__01_T1.tar) means
 LT05-Landsat 5
 L1TP is sensor name, information
 And there is a date that the image is captured and the band number also given.

 Go to raster>>spectral>>layer stack,
 Input the individual file one by one and locate the output file.

6|Page
 file>>open>>raster layer>>open the destination image

 Go to content>>right click the stacked image>>fit layer to window.(A)

 This is the final output which is layer stacked image.(B)

(A) (B)

7|Page
 CHAPTER THREE (BASIC FUNCTION)
Add new viewer
 Home>>add views>>create new 2d views.

Link Viewers
 Open new image in 2nd viewer
 Home>>link viewers>>link views>>control zoom in one viewer and show the exact location in another viewer.

8|Page
Metadata

 Home>select the image>metadata>projection>see the projection of the image

Projection is a method by which the curved surface of the earth is portrayed on a flat surface. This generally
requires a systematic mathematical transformation of the earth's graticule of lines of longitude and latitude
onto a plane.
 If you need re projection , then raster>>re project >>select the categories such as WGS 84,north,UTM 46.

9|Page
Raster Option
 Right click on mouse>>Open raster layer>>select image>>raster option>change the band combination

 Add new viewer>>add raster layer>>open output file

10 | P a g e
 CHAPTER FOUR (MULTISPECTRAL BAND COMBINATION& IMAGE
ENHANCEMENT)

 Go to multispectral and change sensor,band combination such as true color 3,2,1 and false color
4,3,2.

 Manually one can change the combination or use stablished functional system.
 Go to adjust radiometry and changes various stretches.

11 | P a g e
 Go to breakpoint and change them for better quality

 Change Discrete DRA, contrast and also brightness

12 | P a g e
 CHAPTER FIVE(SUBSET)

Shape file method

 Right click>open vector layer(a)

 Select the shape file>open it(b)

(A) (B)
 Select the shape file>vector dialog>re projection
 Re project the shape file as same as the layer stacked image projection

13 | P a g e
 Open the projected shape file>select the shape file>home>copy(a)>paste(b)

(a) (b)

 After it there seems a AOI file automatically add in the contents>>right click>>save layer as>>save
the AOI file

14 | P a g e
 Then raster>>subset and chip>>create subset image>>input the image and select output location
 Select AOI file from the destination folder

 Then open raster layer>>open the subset image

15 | P a g e
Drawing Method
 Drawing>>insert geometry>> select polygon

 Draw polygon on the layer stacked image>>then automatically add a AOI file in contents>>save layer
as>>save the AOI fi

16 | P a g e
 Raster>>subset & chip>>create subset image>>select image>>select output location>>select AOI file

 This is the subset image

17 | P a g e
Subset chip method
 Multispectral>>subset & chip>>place the box in exact location>>select location of the output

 This is the final output

18 | P a g e
 CHAPTER SIX (IMAGE PREPROCESSING)

Correction:
Image processing is a process which makes an image interpretable for a specific use. There are
many methods, but only the most common will be presented here.

radiometric

correction Atmospheric

Geometric

Radiometric Correction
Procedures that correct or calibrate aberrations in data values due to specific distortions from such things
as noise reduction or instrumentation errors (such as striping) in remotely sensed data.

Noise reduction
Applies an edge-preserving smoothing techniques.
 Raster>>Radiometric>>noise reduction>>input file>>select the output location

19 | P a g e
 Open raster layer>>open corrected image>>compare

Brightness Inversion

This dialog allows both linear and non linear of the image intensity range.Dark details become light and
light details become dark.
 Raster>>radiometric>>brightness inversion>>input and output file in selected location

20 | P a g e
 Open raster layer>>corrected photo

Histogram Equalization

Apply a nonlinear contrast stretch that redistributes pixel values so that there are approximately the same
number of pixels with each value within a range.
 Raster>>radiometric >>histogram equalization>>input and output file and location.

 Open raster layer>>open corrected image>> go to metadata>>histogram

 Go to metadata of previous layer- stacked image>>histogram
 Then compare the two images

21 | P a g e
 Here the corrected image histogram show the largely distributed pixel value

Atmospheric Correction
Haze reduction
Atmospheric effects can cause imagery to have a limited dynamic range, appearing as haziness or
reduced contrast. Use this dialog to sharpen an image using Tasseled Cap or Point Spread Convolution.
For multispectral images, this method is based on the Tasseled Cap transformation that yields a
component that correlates with haze. This component is removed and the image is transformed back
into RGB space. For panchromatic images, an inverse point spread convolution is used.
 Radiometric>>haze reduction>>input and output the file >>open new raster layer(a)
 open corrected file(b)

(A) (B)

22 | P a g e
 (a) (b)

 Compare and see the changes

Geometric Correction
Mosaic

Use MosaicPro workstation to join georeferenced images and form a larger image or a set of images (these
mosaicked project files are named with an .mop file extension). The input images must all contain map and
projection information and have the same number of layers. They do not need to be in the same projection or
have the same pixel cell sizes. Calibrated input images are also supported.
 Go to raster>>Mosaic pro>>display add images dialog>>add images

 Then again add image through same process

23 | P a g e
 Click process in the bar and run mosaic>>select output folder
 Open raster layer>>add mosaic image

In EarthExplorer maximum satellite images are geometrically corrected.So it is not need to do

geometric correction in details.
24 | P a g e
 CHAPTER SEVEN (IMAGE INTERPRETATION)
Inquire Tools
 Home>>select>>inquire tools>>open image>>select feature in map
 Then evaluate the pixel value with band combination.

 In Landsat 5 ,false color composition 4,3,2 is a popular combination where red indicate
vegetation ,blue indicates water.so in the cursor is in waterbody the pixel value of blue is larger.

25 | P a g e
Measurement

Click to measure position, length, direction, area, and so forth, and optionally save the measurements to
an annotation layer.Measurements can be based on a Cartesian system or on the ellipsoid.

 Home>>measure>>select parameter such as sq. Kilometer, meter etc.(a)

 select point>>polyline>>Set the line on a feature or objectives and measure(b)

(a)

(b)

26 | P a g e
 Select point>>polygon>>draw a polygon of a feature and then measure.The measurement dialog
show it.

NDVI
This is done in order to suppress, or normalize, varying effects such as viewing angles, sun shading,
atmospheric effects, soil difference, and so on. It is also applied to maximize sensitivity to the feature of
interest, such as the relative health of vegetation. To achieve this most Indices go beyond simple band
division to include differencing, weighting, and the introduction of other variables
 Go to raster>>unsupervised>>NDVI

27 | P a g e
 Select the sensor such as Landsat Tm and make sure that NIR and RED bands are selected
 Select input and output file

 Home>>inquire tools>>placed on this on ndvi image.

 It measure the file pixel and the range is +1>0>-1 .

 Less than zero means there is no vegetation

28 | P a g e
Profile Tools
Spectral Profile

Use the Spectral Profile dialog to visualize the reflectance spectrum of a single pixel through many bands.
This technique is particularly useful for hyperspectral data that can have hundreds of layers. This technique
allows estimates of the chemical composition of the material in the pixel. You can compare the profiles that
you generate to those from laboratory (or field) spectrophotometers.

 Go to multispectral>>spectral profile

 Select the inquire tools and plot the features.It shows the pixel value for different layer.

29 | P a g e
Surface Profile
Use the Surface Profile Viewer to visualize the reflectance spectrum of a rectangular area of data file
values in a single band of data. You can overlay the wireframe surface with a grayscale, thematic, or true
color image.

 Go to multispectral>>spectral>>surface profile

 Select feature and plot different layers.It shows 3d model of pixel values of different features.

30 | P a g e
Spatial Profile
Use the Spatial Profile Viewer to visualize the reflectance spectrum of a polyline of data file values in a
single band of data (one-dimensional mode) or in many bands (perspective three-dimensional mode).
The most common example of single band data profile is that of a Digital Elevation Model (DEM) being
used to create a height cross-section profile along a route. This helps in interpreting changes in elevation
along a planned route and in identifying sections of the route which are particularly steep or flat.

 Go to multispectral>>surface profile>>spatial profile

 It shows pixel values with distance that you draw, in a plotted layer.

31 | P a g e
Unsupervised Classification

ERDAS IMAGINE uses the ISODATA algorithm to perform an unsupervised classification. ISODATA
stands for Iterative Self-Organizing Data Analysis Technique. It is iterative in that it repeatedly performs
an entire classification (outputting a thematic raster layer) and recalculates statistics. Self-Organizing refers
to the way in which it locates the clusters that are inherent in the data.

 Go to raster>>unsupervised>>unsupervised classification>>input and output files

destination>>define classes in isodata>>define iteration, means scanning the main image

 Open raster layer and open that unsupervised image.

 Go to>>table>>show attribute>>edit class name,color (click on the icon) and also opacity by clicking
on it

32 | P a g e
 CHAPTER EIGHT (MAP PRODUCTION)

Supervised Classification
Supervised Classification provides tools for categorizing pixels using interactive supervised techniques.
You provide examples of what particular classes look like, which are then used by the software algorithms
to derive rules for mapping all other pixels into the class values.It is done by manually.

 Go to raster>>supervised>>signature editor>>
 Home>>drawing tools>>create polygon and add them as a class in signature features by clicking create
new signature as AOI>>again do it of same feature but in another location.
 When many classes of one feature added, merge them(a)
 make another one class automatically ,delete the previous selection.(b)

(A)

33 | P a g e
 (b)
 Again create polygon of another feature by drawing tools and add them in signature tools as class like
previous method.
 doing it one by one and add classes in signature tools.
 Edit the class name and color also by click it such as water-blue color
 After completing manually drawing method, save signature file as .sig.

 Then go to supervised classification>>locate input,output and .sig files.(a)

 After doing that open raster layer>>open the supervised image.(b)
 Go to home>>table>>open attribute table
34 | P a g e
 (a)

(b)

It is done by manually when one can go to the location and analyses the features and observe them.Finally
show the analyzed features on map by different classes but in unsupervised classification software done it
automatically by stablished algorithm.

35 | P a g e
Unsupervised Classification
 See details in page no-32
 There is a unsupervised classification with 5 classes such as water,bare
land,vegetation,crop,settlement. Identify them with different colors in map

 Add views>>reate new map view

 There add a viewer,click on map view,go to layout >>map frame
 Select map template and edit the map tile also

36 | P a g e
 Double click on map viewer,after adding map frame,make sure that there is another viewer where
unsupervised image is opened.Placed the unsupervised image on the box and click ok of the map
frame tools.

 The unsupervised map is open on the map viewer

37 | P a g e
  Layout>>map grid>>map grid
 Change margin>>length outside as your wish

38 | P a g e
 Add the scale bar properties and select parameter such as meter.

 Select Legend option and identify the classes.

 Modify yourself

39 | P a g e
 Add the north line
 Add locational map also
 Final map is ready
 File>>print

THE END

40 | P a g e