B.

Tech AGRICULTURE ENGINEERING

AI3701 - REMOTE SENSING & GIS
UNIT - 2 DATA PRODUCTS AND IMAGE ANALYSIS
PART - A
1) What are the types of data products?
Based on level of processing:
● standard products
● value added products and
● derived products
Based on output media/ scale:
● photographic media
● digital media
Based on area of coverage:
● Full Scenes-all bands
● Full Scenes-specified bands
2) What is image interpretation?
Image interpretation is defined as the extraction of qualitative and quantitative
information in the form of a map, about the shape, location, structure, function, quality,
condition, relationship of and between objects, etc. by using human knowledge or experience.
3) Differentiate supervised & unsupervised classification.
A supervised classification algorithm requires a training sample for each class i.e a
collection of data points known to have come from the class of interest. The classification is
based on how near a point to be classified is to the training data set.
Unsupervised classification algorithms do not compare points to be classified with
training data. The unsupervised algorithms examine a large number of unknown data vectors
and divide them into classes based on properties inherent to the data themselves. Thus the
classes are separated from the differences observed within the data.
4) What is the data for all types of sensors?
The data for all the sensors of IRS -1C/1D are supplied on digital media like
a) Computer compatible tapes (CCTs)
b) Cartridge tapes
c) Floppies
d) CD-ROM products
5) List the image enhancement techniques
● Contrast manipulation.
 ● Spatial feature manipulation.
● Multistage manipulation.
6) What is multi spectral image classification?
● Supervised Classification techniques &
● Unsupervised Classification.
7) Define visual interpretation keys.
Visual interpretation is the act or process of examining images (satellite imageries) for the
purpose of identifying objects and assessing their significance. Visual image interpretation
involves detection, recognition, identification, classification and delineation of objects in an
aerial or a satellite image.
8) What is an image interpretation strategy?
Image processing is a physical process used to convert an image signal into a physical
image. Image processing usually refers to digital image processing, but optical and analog
image processing also are possible.
9) Describe an image with its properties.
Image processing is a physical process used to convert an image signal into a physical
image. Image processing usually refers to digital image processing, but optical and analog
image processing also are possible.
10) What is visual image interpretation?
Visual interpretation is the act or process of examining images (satellite imageries) for
the purpose of identifying objects and assessing their significance. Visual image
interpretation involves detection, recognition, identification, classification and delineation of
objects in an aerial or a satellite image.
11) What are the two approaches of image classification?
1. Supervised Classification
2. Unsupervised Classification
12) What are the errors arising through processing ?
Data processing errors, algorithmic errors, calibration errors,data fusion errors,
interpolation errors, geometric errors,temporal errors, statistical errors, software and
hardware errors, human errors.
13) Define Interpretation keys and its types.
● It is the criteria for identification of an object with interpretation elements.
● Image-interpretation depends on the interpretation keys which an experienced
interpreter has established from prior knowledge and the study of the current images.
● Basically, interpretation keys can be one of two generic types:
 • Selective keys
• Elimination keys.
14) Define preprocessing.
In the preprocessing, prepare data for subsequent analysis that attempts to correct or
compensate for systematic errors.
15) Write short notes on image enhancement.
The operations are carried out to improve the interpretability of the image by increasing
apparent contrast among various features in the scene. The enhancement techniques depend
upon two factors mainly l The digital data (i.e. with spectral bands and resolution)
16) List the steps of image processing.
The three steps of image processing are,
• Pre-processing
• Display and enhancement
• Information extraction
17) What are image processing & Image processing techniques?
Image processing is a physical process used to convert an image signal into a physical
image. Image processing usually refers to digital image processing, but optical and analog
image processing also are possible.
Techniques:
1. Preprocessing
2. Image Enhancement
3. Image Classification
4. Image Registration
5. Feature Extraction
6. Change Detection
7. Geospatial Analysis
8. Object
18) What is a geo coded product?
1. A geo-coded product in remote sensing provides specific geographic coordinates
(latitude and longitude) for each pixel or data point in an image or dataset.
2. This georeferencing is vital for precise spatial analysis and integration with other
geospatial data, enhancing applications such as land use mapping and disaster management.
19) Explain value added products and its types?
● When standard products are processed according to specifications and requirements of
users, these products get converted into value added products.
 ● These products are basically of four types viz.
• Geocoded products
• Merged products
• Ortho product
•Template registered products.
20) Explain path number or track number.
The pole to pole coverage on the earth for each pass of the satellite is given a specific
number called path number or track number.
21) How are along track and across track scanners different?
There are two methods of scanning used to obtain multi spectral image data. They are
along track and across track scanners. Across Track Scanner Along Track Scanner.
1. They scan earth in a series of lines They also scan earth in a series of lines.
2. The lines are oriented perpendicular to the direction of motion of the sensor platform and
to the swath. The lines are oriented perpendicular to the flight direction.
22) What are the types of image interpretation?
Image be classified by adopting either of the following methods
1. Spectral pattern recognition
2. Spatial pattern
3. Supervised classification
4. Unsupervised classification
23) What are the image interpretation elements?
The eight elements of image interpretation are shape, size, tone, shadows, texture, site,
pattern and association.
24) Define digital image processing.
Digital Image Processing is a collection of techniques for the manipulation of digital
images by computers. The raw data received from the imaging sensors on the satellite
platforms contains flaws and deficiencies. To overcome these flaws and deficiencies in order
to get the originality of the data, it needs to undergo several steps of processing. This will
vary from image to image depending on the type of image format, initial condition of the
image and the information of interest and the composition of the image scene.
25) What is the data for all types of sensors?
Sensors collect various types of data depending on their purpose, including temperature,
pressure, light intensity, proximity, humidity, acceleration, angular velocity, magnetic field
strength, GPS coordinates, infrared radiation, sound intensity, colour information, biometric
data, gas concentrations, motion, images, distance, force, orientation, and radiation levels.
 PART - B
1) Write about a supervised classification system and explain any one of the
classification.
1. Supervised Classification
● In this classification method, an analyst identifies the imaginary in terms of
homogeneous representative samples of different surface cover types of interest.
● These samples are called “Training Areas”.
● The selection of appropriate training areas is based on the
analyst’s familiarity with the geographical area and
knowledge of the actual surface cover types present in the
image.
● The numerical information in all spectral bands for the
pixels comprising these areas are used to train the computer
to recognize specially similar areas for each class.
● Therefore, in supervised classification, the analyst first
identifies the information classes based on which it
determines the spectral classes which represent them.
● Supervised classification requires the analyst to select
training areas where he/she knows what is on the ground
and then digitise a polygon within that area.

Common classifiers:
1. Parallel piped classifiers
 2. Minimum distance to mean classifiers
3. Maximum likelihood classifiers (MLC)
1)Parallelepiped Classifiers:
• In this method, a parallelepiped-like (hyper rectangle)
subspace is defined for each class.
• Using the training data for each class, the limits of a
parallelepiped subspace can be defined by the minimum and
maximum pixel values in the given class.
• The pixels lying inside the parallelepipeds are tagged to
this class.
(Or)
2) Minimum distance to mean classifiers:
• Each cluster can be represented by its centroid, often
its mean value.
• The multidimensional distance to each cluster centroid
is calculated, and the pixel is then assigned to the closest
cluster
• Not generally used for remote sensing purposes.
(Or)
3) Maximum likelihood classifiers:
● Most common supervised method used in Remote sensing.
● The probability of a pixel belonging to each of the predefined classes is calculated
and the pixel is assigned to the class with the highest probability.
2) Write about elements of visual interpretation with good examples.
VISUAL INTERPRETATION ELEMENTS
1. Tone
2. Shape
3. Size
4. Pattern
5. Texture
6. Shadow
7. Association
1. TONE
● Tone refers to the relative brightness or colour of objects in an image.
● Generally, tone is the fundamental element for
distinguishing between different targets or
features.
● Variations in tone also allows the elements of
shape, texture, and pattern of objects to be
distinguished.

2. SHAPE
● Shape refers to the general form, structure,
or outline of individual objects.
● Shape can be a very distinctive clue for
interpretation.
● Straight edge shapes typically represent
urban or agricultural (field) targets, while
natural features, such as forest edges, are
generally more irregular in shape, except
where man has created a road or clear cuts.
● Farm or crop land irrigated by rotating sprinkler systems would appear as circular
shapes.

3. SIZE
● Size of objects in an image is a function of scale.
● It is important to assess the size of a target relative to
other objects in a scene, as well as the absolute size,
to aid in the interpretation of that target.
● A quick approximation of target size can direct
interpretation to an appropriate result more quickly.
● E.g. If an interpreter had identified an area with a
number of buildings in it, large buildings such as
factories or warehouses would suggest commercial
property, whereas small buildings would indicate residential use.
4. PATTERN
● Pattern refers to the spatial arrangement of
visibly discernible objects.
● Typically an orderly repetition of similar tones
and textures will produce a distinctive and
ultimately recognizable pattern.
● E.g. Orchards with evenly spaced trees, and
urban streets with regularly spaced houses.

5. TEXTURE

● It refers to the arrangement and frequency of tonal

variation in particular areas of an image.
● Rough textures would consist of a mottled tone
where the grey levels change abruptly in a small
area, whereas smooth textures would have very
little tonal variation.
● Smooth textures are most often the result
uniform, even surfaces, such as fields, asphalt,
or grasslands.
● A target with a rough surface and irregular structure, such as a forest canopy, results
in a
rough textured appearance.

6. SHADOW
● Shadow is also helpful in interpretation as it maps
provide an idea of the profile and relative height of a
target or targets which may make identification easier.
● However, shadows can also reduce or eliminate
interpretation in their area of influence, since targets
within shadows are much less discernible from
their surroundings.
● Shadow is also useful for enhancing or identifying
topography and landforms,particularly in
 radar imagery.

7. ASSOCIATION
● Association takes into account the relationship
between other recognizable objects or features in
proximity to the target of interest.
● The identification of features that one would expect
to associate with other features may provide
information to facilitate identification.
● E.g. Commercial properties may be associated with
proximity to major transportation routes,whereas
residential areas would be associated with
schools,playgrounds, and sports fields.
● In our example, a lake is associated with boats, a marina,
and adjacent recreational land.
3) Explain various image enhancement and image classification techniques.
IMAGE ENHANCEMENT
● Low sensitivity of the detector, weak signal of the objects present on the earth
surface, similar reflectance of different objects and environmental conditions at the
time of recording are the major causes of low contrast of the image.
● Human eye is poor at discriminating against the slight spectral or radiometric
differences that may characterise the feature.
● Image enhancement techniques improve the quality of an image as perceived by the
human eye.
● The principle of image enhancement is the process of a given image so that the result
is more suitable than the original image for a specific application.
● It accentuates or sharpens image features such as edges, boundaries, or contrast to
make a graphic display more helpful for display and analysis.
● The enhancement doesn't increase the inherent information content of the data, but it
increases the dynamic range of the chosen features so that they can be detected easily.
● There exist a wide variety of techniques for improving image quality. Contrast stretch,
density slicing, edge enhancement, and spatial filtering are the most commonly used
techniques.
● Image enhancement method are applied separately to each band of a multispectral
image
Point Operations
● Modification of brightness values of each pixel in an image data set independently.
(radiometric enhancement)
● Brings out contrast in the image.
Local operation
Modification of pixel values based on the values of surrounding pixels. (spatial
enhancement)
Image Transformations
Enhancing images by transforming the values of each pixel on a multiband basis (spectral
enhancement )
i) Contrast enhancement:
● Low-contrast images can result from poor illumination, lack of dynamic range in the
image sensor, or even wrong setting of a lens aperture during image acquisition.
● The idea behind contrast stretching is to increase the dynamic range of the grey levels
in the image being processed.
ii) Linear stretch:
● Rescaling the original brightness values proportionately into a new distribution based
on a specified minimum and maximum value.
● A DN in the low range of the original histogram is assigned to extreme black, and a
value at the high end is assigned to extreme white.
● The remaining pixel values are distributed linearly between these two extremes
iii) Histogram equalisation:
● In this technique, histogram of the original image is redistributed to produce a
uniform population density.
● This is obtained by grouping certain adjacent grey values.
● Thus the number of grey levels in the enhanced image is less than the number of grey
levels in the original image.
● Contrast is increased at the most populated range of brightness values of the
histogram (or "peaks").
● It automatically reduces the contrast in very light or dark parts of the image
associated with the tails of a normally distributed histogram
iv) Density Slicing:
● Mapping intervals of brightness values into discrete colours.
● Grey-level slicing: Highlighting a specific range of grey levels in an image often is
desired.
 ● Applications include enhancing features such as masses of water in satellite imagery
and enhancing flaws in x-ray images.
v) Edge Enhancement/Local enhancement:
● It is often necessary to enhance details over small areas.
● Enhancing contrast in a local region to reinforce visual transitions between areas of
contrasting brightness.
IMAGE CLASSIFICATION
● Image Classification is a procedure to automatically categorise all pixels in an image
of a terrain into land use and land cover classes.
● This concept is dealt with under broad subject, namely,“pattern recognition”.
● Spectral pattern recognition refers to the family of classification procedures that
utilises this pixel-by-pixel spectral information as the basis for automated land cover
classification.
● Spatial pattern recognition of image on the basis of the spatial relationship with the
pixel surrounding.
Image classification technique are grouped into two types, namely,
1. Supervised Classification
2. Unsupervised Classification.
1. Supervised Classification
● In this classification method, an analyst identifies the imaginary in terms of
homogeneous representative samples of different surface cover types of interest.
● These samples are called“Training Areas”.
● The selection of appropriate training areas is based on the
analyst’s familiarity with the geographical area and
knowledge of the actual surface cover types present in the
image.
● The numerical information in all spectral bands for the
pixels comprising these areas are used to train the computer
to recognize specially similar areas for each class.
● Therefore, in supervised classification, the analyst first
identifies the information classes based on which it
determines the spectral classes which represent them.
● Supervised classification requires the analyst to select
training areas where he/she knows what is on the ground
and then digitise a polygon within that area.
Common classifiers:
1. Parallel piped classifiers.
2. Minimum distance to mean classifiers.
3. Maximum likelihood classifiers (MLC)
2. Unsupervised Classification
● The Unsupervised Classifiers do not utilise training data as a basis for classification.
● These classifiers examine the unknown pixel and aggregate them into a number of
classes based on natural grouping or clusters present in the image values.
● The classes that result from unsupervised classification are spectral classes.
● The analyst must compare the classified data into some form of reference data to
identify the informational value of the spectral classes
4) Write in detail about different types of data products.
Data products:
Data products are data from various sensors in a suitable and standard form and format,
which can be readily used by users.
TYPES OF DATA PRODUCTS
Based on level of processing:
● standard products
● value added products and
● derived products
Based on output media/scale:
● photographic media
 ● digital media
Based on area of coverage:
● Full Scenes-all bands
● Full Scenes-specified bands
● Full Scenes with Shift-Along Track
● Quadrants
● Geocoded Floating (15’x 15’)
● Geocoded (7½’x7½’) map sheet/floating
● Full pass/strip data of one array
BASED ON LEVEL OF PROCESSING:
A)Standard products:
Generated using information received directly from the satellites and by applying necessary
radiometric and geometric corrections.
a) Path-Row Based Standard Products:
b) Shift Along Track (SAT) Products
c) Quadrant Products
d) Georeferenced Products
e) Basic Stereo Products
f) Area of Interest (AOI) based Products
a) Path-Row Based Standard Products:
● Actually, the data are recorded by the sensors along the track/path of the satellite
covering a specific width (across the satellite track/path).
● The user specifies path and row of the scene, sensor,sub-scene (if any), number of
scenes, date of pass of the satellite, band numbers/band combination (for
photographic products) and the product code (specified by the data provider) as
necessary information to procure the desired products from the provider.
b) Shift Along Track (SAT) Products:
● In case a user’s area of interest falls between two successive scenes of the same path
(one below the other), the data can be supplied by sliding the scene in along-the-track
direction.
● These products are called Shift-Along-Track Products. In this case, the percentage
(10% to 90% in multiples of 10%) of shift has to be specified by the user in addition
to the other necessary information about both the scenes.
c) Quadrant Products:
 ● In IRS series of satellites, LISS III full scene (path/row based products) data has been
divided into 4 nominal quadrants designated by letter A, B, C & D.
● Each of these quadrants is one full scene of Payload Panchromatic camera (PAN)
data (B & W data), and is further divided into 9 quadrants, given numbers from 1 to 9
which are nothing but PAN sub-scenes.

d) Georeferenced Products:
● Geo referencing is the process of transforming the remote sensing data or a map to a
coordinate and projection system.
● This means, all the objects/elements in remote sensing data and the map get specific
geographic location in terms of longitudes and latitudes.
● Thus, georeferenced products are north-east oriented products.
e) Basic Stereo Products:
● A stereo-pair comprises two images of the same area, acquired on different dates or
same day from different angles.
● These products are used for generating digital elevation maps or 3D visualisation of
the study area.
● Cartosat-1 mission provides along track stereo images.
f) Area of Interest (AOI) based Products:
● Users may define their area of interest according to the extent of study area.
● Scenes covering the user’s AOI can be selected and ordered for procurement.
● The scenes covering the users AOI are provided as tiles and are not mosaicked.
● The minimum order/quantity for IRS P5 products is 25 × 25 sq km and that for IRS-
P6 data is 23 × 23 sq km for LISS IV and maximum order/quantity is 10,000 sq km.
● AOI based products are provided as digital products (CD/DVD) only.
B) VALUE ADDED PRODUCTS
When standard products are processed according to specifications and requirements of users,
these products get converted into value added products.
These products are basically of four types viz.
• Geocoded products
• Merged products
• Ortho product
• Template registered products.
a) Geocoded Products:
● These products are also known as georeferenced products (mentioned in one of the
preceding paragraphs).
b) Merged Products:
● Remote sensing satellites record Panchromatic/mono-band data (higher spatial
resolution) and multi-band data (low in spatial resolution).
● When MSS data area is merged with panchromatic data, we get MSS data with spatial
resolution of panchromatic data.

c) Ortho Products:
 ● Geometrically corrected products, with corrections for displacement caused by tilt
and relief, are called Ortho Products.
● Therefore, ortho images show ground objects in their true planimetric positions, like
the positions of objects in a map.
d) Template Registered Products:
● In crop monitoring, data of the same area having similar geometric fidelity and
temporally registered are required-Template registered products.
● E.g.: Data from AWiFS (Advanced Wide Field Sensor) sensor.
C) DERIVED PRODUCTS
● Information extracted from remote sensing data is known as derived data products.
● E.g. Vegetation index map or sea surface temperature profiles extracted from various
remote sensing data products are called derived products.
● These derived products are generated by further processing /analysing the data, and
are readily usable by the user.
BASED ON OUTPUT MEDIA/SCALE:
a) Photographic media
b) Digital media.
● Photographic products are supplied on film or paper prints.
● The scale of these photographic products can range from 1:1M to 1:5000.
5) Explain briefly about the image enhancement techniques.
IMAGE ENHANCEMENT
● Low sensitivity of the detector, weak signal of the objects present on the earth
surface, similar reflectance of different objects and environmental conditions at the
time of recording are the major causes of low contrast of the image.
● Human eye is poor at discriminating against the slight spectral or radiometric
differences that may characterise the feature.
● Image enhancement techniques improve the quality of an image as perceived by the
human eye.
● The principle of image enhancement is the process of a given image so that the result
is more suitable than the original image for a specific application.
● It accentuates or sharpens image features such as edges, boundaries, or contrast to
make a graphic display more helpful for display and analysis.
● The enhancement doesn't increase the inherent information content of the data, but it
increases the dynamic range of the chosen features so that they can be detected easily.
 ● There exist a wide variety of techniques for improving image quality. Contrast stretch,
density slicing, edge enhancement, and spatial filtering are the most commonly used
techniques.
● Image enhancement method are applied separately to each band of a multispectral
image
Point Operations
● Modification of brightness values of each pixel in an image data set independently.
(radiometric enhancement)
● Brings out contrast in the image.
Local operation
Modification of pixel values based on the values of surrounding pixels. (spatial
enhancement)
Image Transformations
Enhancing images by transforming the values of each pixel on a multiband basis (spectral
enhancement )
i) Contrast enhancement:
● Low-contrast images can result from poor illumination, lack of dynamic range in the
image sensor, or even wrong setting of a lens aperture during image acquisition.
● The idea behind contrast stretching is to increase the dynamic range of the grey levels
in the image being processed.
ii) Linear stretch:
● Rescaling the original brightness values proportionately into a new distribution based
on a specified minimum and maximum value.
● A DN in the low range of the original histogram is assigned to extreme black, and a
value at the high end is assigned to extreme white.
● The remaining pixel values are distributed linearly between these two extremes
iii) Histogram equalisation:
● In this technique, histogram of the original image is redistributed to produce a
uniform population density.
● This is obtained by grouping certain adjacent grey values.
● Thus the number of grey levels in the enhanced image is less than the number of grey
levels in the original image.
● Contrast is increased at the most populated range of brightness values of the
histogram (or "peaks").
 ● It automatically reduces the contrast in very light or dark parts of the image
associated with the tails of a normally distributed histogram
iv) Density Slicing:
● Mapping intervals of brightness values into discrete colours.
● Grey-level slicing: Highlighting a specific range of grey levels in an image often is
desired.
● Applications include enhancing features such as masses of water in satellite imagery
and enhancing flaws in x-ray images.
v) Edge Enhancement/Local enhancement:
● It is often necessary to enhance details over small areas.
● Enhancing contrast in a local region to reinforce visual transitions between areas of
contrasting brightness.
6) Explain in detail about image interpretation.
Image interpretation is the process of extracting meaningful information from images. There
are several types of image interpretation, each with its own purpose and methods. Here are
some of the main types:
1. Visual Interpretation.
Qualitative Analysis: Visual interpretation involves the human eye and brain to
perceive and analyse images. It's often used for tasks like land cover classification,
where an analyst visually identifies and classifies objects or features in an image.
Photo Interpretation: This is a specialised form of visual interpretation used for aerial
and satellite imagery. Analysts use stereoscopes to view images in 3D, aiding in the
interpretation of terrain and objects.
2. Digital Image Processing:
Enhancement: Digital image processing techniques are applied to improve the quality
of images. This includes techniques like contrast enhancement, noise reduction, and
sharpening to make features more discernible.
Image Classification: Supervised or unsupervised machine learning algorithms can be
used to classify pixels or regions in an image into predefined categories. This is
commonly used in remote sensing applications.
3. Multispectral and Hyperspectral Analysis:
Multispectral: In remote sensing, images are captured in multiple spectral bands.
Analysts can interpret these images to distinguish various land cover types, monitor
vegetation health, or identify mineral deposits.
 Hyperspectral: These images capture a large number of very narrow spectral bands.
This allows for detailed material identification, including differentiating between
vegetation species or detecting specific chemicals.
4. Thermal Imaging Interpretation:
Thermal infrared imagery: Used for detecting variations in temperature.
Applications include identifying heat leaks in buildings, monitoring wildfires, and
locating warm-blooded animals in the wild.
5. Medical Image Interpretation:
Diagnostic Imaging: In medicine, images like X-rays, MRI, and CT scans are
interpreted by medical professionals to diagnose diseases, injuries, or abnormalities.
Computer-Aided Diagnosis (CAD): Machine learning algorithms assist radiologists
in detecting anomalies in medical images, improving accuracy.
6. Satellite and Aerial Image Interpretation:
Urban Planning: Satellite and aerial images are used for urban planning, including
identifying land use patterns, infrastructure planning, and disaster assessment.
Environmental Monitoring: Interpretation of satellite imagery helps in monitoring
environmental changes such as deforestation, pollution, and coastal erosion.
7. Geological Image Interpretation:
Mineral Exploration: Geological remote sensing uses images to identify geological
structures, mineral deposits, and potential mining sites.
Landform Analysis: Interpretation of topographic and geological maps aids in
understanding landforms and subsurface features.
8. Security and Surveillance:
Object Recognition: Image interpretation is used in security and surveillance systems
to detect and identify objects, faces, or suspicious activities.
Anomaly Detection: Algorithms can automatically detect unusual patterns or objects
in video feeds, helping with security.
9. Astronomical Image Interpretation:
Astronomical Imaging: Interpretation of astronomical images helps astronomers study
celestial objects, discover new phenomena, and understand the universe's properties.
Each type of image interpretation requires specific knowledge, tools, and techniques.
Advancements in artificial intelligence, machine learning, and computer vision have
expanded the capabilities and accuracy of automated image interpretation in recent years,
making it a crucial component in various fields.