RS GIS UNIT -1

UNIT – I Concepts of Remote Sensing Basics of remote sensing- elements involved in

remote sensing, electromagnetic spectrum, remote sensing terminology & units, energy
resources, energy interactions with earth surface features & atmosphere, atmospheric
effects, satellite orbits, Sensor Resolution, types of sensors. Remote Sensing Platforms and
Sensors, IRS satellites.

Remote Sensing Data Interpretation Visual interpretation techniques, basic elements,

converging evidence, interpretation for terrain evaluation, spectral properties of soil, water
and vegetation. Concepts of Digital image processing, image enhancements, qualitative &
quantitative analysis and pattern recognition, classification techniques and accuracy
estimation.

Overview of GIS and Remote Sensing

Introduction to GIS and Remote Sensing
Geographical Information Systems (GIS) and Remote Sensing are tools used for
understanding and analyzing the Earth's surface and atmosphere. Geographical information
refers to data tied to specific locations, typically on the Earth's surface, but can also apply to
other spatial frames like the human body (in medical imaging) or buildings.
Spatial Data: While ‘geographical’ is commonly used, ‘spatial’ is often preferred in modern
contexts. It refers to information associated with particular locations, whether on Earth or
within other systems.
What is GIS?
GIS is the integration and analysis of geographical data in digital form. The term has evolved
to encompass any manipulation of geographical data using computer systems, including:
Map Making: Using computers to create maps.
Geospatial Analysis: Analysing geographical information to identify patterns and
relationships.
Forecasting: Using complex models to predict future geographical events or changes.
GIS involves more than simply processing raw data; it includes the ability to analyze and
interpret the data to support decision-making in fields such as urban planning,
environmental management, and disaster response.
Role of Remote Sensing
Remote sensing refers to the collection of geographical data, often in the form of satellite
images, without direct contact with the observed area. These satellite images provide
valuable data on the Earth's surface and atmosphere, such as:
Earth Observation: Collecting images of land, water, and vegetation.
Environmental Monitoring: Tracking changes in ecosystems, climate, and land use.
However, systems that process these remote sensing images are not always classified as GIS
unless they integrate or work with additional geographical data types.
GIS vs. Remote Sensing
Remote Sensing: Focuses on collecting geographical data through sensors, typically from
satellites or aircraft.

1
GIS: Involves integrating remote sensing data with other geographical data sources and
conducting analyses on this combined data.
GIS systems are generally more complex than those used for remote sensing because they
process and integrate multiple data types for analysis, forecasting, and decision-making.
Applications in Various Fields
Multidisciplinary Work: GIS is widely used in studies that involve multiple disciplines, such
as environmental science, urban planning, agriculture, and policy development.
Specialized Software: For example, atmospheric scientists or oceanographers use GIS in
ways that are more focused on their specific field, often integrating data from remote
sensing with discipline-specific models.
In summary, GIS and Remote Sensing are complementary technologies used to understand,
analyze, and model the physical world. While remote sensing provides raw data, GIS
integrates and analyzes that data to generate useful insights for various applications.

Concept of Remote Sensing

Definition of Remote Sensing
Remote sensing is defined as the science and art of obtaining information about an object,
area, or phenomenon through the analyses of data acquired by the sensor that is not in
direct contact with the target of investigation
The primary source of data in remote sensing is electromagnetic radiation reflected or
emitted by an object. However, other media such as gravity or magnetic fields can also be
used for remote sensing.
Key Components of Remote Sensing
Remote Sensor: A device used to detect electromagnetic radiation reflected or emitted from
an object. Common examples include cameras and scanners.
Platform: The vehicle that carries the sensor. Aircraft and satellites are typical platforms for
remote sensing.
HISTORY OF REMOTE SENSING:
1839 - First photograph
1858 - First photo from a balloon
1903 - First plane
1909 - First photo from a plane
1903-34 -B/W infrared film WW I and WW II
1960 – Space

2
Energy Source or Illumination
Radiation and the Atmosphere
Interaction with the Target
Recording of Energy by the Sensor
Transmission, Reception, and Processing
Interpretation and Analysis
Application

Energy Source or Illumination (A) – the first requirement for remote sensing is to have an
energy source which illuminates or provides electromagnetic energy to the target of
interest.
Radiation and the Atmosphere (B) – as the energy travels from its source to the target, it
will come in contact with and interact with the atmosphere it passes through. This
interaction may take place a second time as the energy travels from the target to the sensor.
Interaction with the Target (C) - once the energy makes its way to the target through the
atmosphere, it interacts with the target depending on the properties of both the target and
the radiation.
Recording of Energy by the Sensor (D) - after the energy has been scattered by,or emitted
from the target, we require a sensor (remote - not in contact with the target) to collect and
record the electromagnetic radiation.
Transmission, Reception, and Processing (E) - the energy recorded by the sensor has to be
transmitted, often in electronic form, to a receiving and processing station where the data
are processed into an image (hardcopy and/or digital).
Interpretation and Analysis (F) - the processed image is interpreted, visually and/or digitally
or electronically, to extract information about the target which was illuminated.
Application (G) - the final element of the remote sensing process is achieved when we apply
the Information, we have been able to extract from the imagery about the target in order to
better understand it, reveal some new information, or assist in solving a particular problem.
Advantages of Remote Sensing
Large Area Coverage: Enables monitoring of extensive regions in a single observation.

3
Global Accessibility: Provides data from inaccessible or remote areas.
Frequent Observations: Captures data repeatedly for monitoring changes over time.
Multispectral and Multitemporal Analysis: Offers insights across various wavelengths and
time frames.
Non-Intrusive: Collects data without disturbing the environment.
Rapid Data Collection: Gathers information quickly compared to ground-based surveys.
Integration with GIS: Enhances decision-making by combining spatial data with geographic
tools.
Versatile Applications: Useful in agriculture, forestry, urban planning, disaster management,
and climate studies.
Disadvantages of Remote Sensing
High Initial Cost: Requires significant investment in satellite systems and sensors.
Complex Data Interpretation: Needs expertise to analyze and process data accurately.
Limited Temporal Resolution: Satellite revisits may not align with specific event timings.
Weather Dependency (Passive Sensors): Cloud cover and weather conditions can obstruct
data collection.
Data Accuracy Limitations: Errors due to sensor calibration, atmospheric effects, or
resolution constraints.
Specialized Equipment Required: High-resolution applications demand advanced and costly
tools.
Ethical Concerns: Potential issues with surveillance and privacy in human habitats.
Dependence on Ground Truthing: Requires on-site validation to ensure data accuracy.
Applications of Remote Sensing
Agriculture
Monitors crop health, soil conditions, and irrigation efficiency.
Supports precision farming and forecasts crop yields.
Forestry
Tracks deforestation, forest fires, and canopy density.
Assists in biodiversity conservation and forest management.
Urban Planning
Maps land-use patterns and urban sprawl.
Aids infrastructure development and sustainable planning.
Disaster Management
Provides real-time data on floods, earthquakes, and cyclones.
Supports risk assessment and post-disaster recovery.
Climate and Weather Studies
Monitors global weather patterns and climate changes.
Collects atmospheric data for forecasting and analysis.
Water Resources Management
Tracks water body levels and river basin dynamics.
Supports flood risk analysis and watershed management.

4
Geology and Mineral Exploration
Identifies mineral deposits and geological structures.
Maps fault lines and supports natural resource exploration.
Coastal and Marine Studies
Monitors coastal erosion, marine ecosystems, and coral reefs.
Assists in sustainable marine resource management.
Environmental Monitoring
Tracks pollution levels, desertification, and glacier melting.
Supports initiatives for environmental conservation.
Transportation and Navigation
Maps transportation networks and routes.
Enhances traffic management and navigation systems.
Defence and Surveillance
Provides strategic data for border security and surveillance.
Assists in military planning and reconnaissance.
Biodiversity Conservation
Maps habitats and tracks wildlife migration.
Aids in planning and protecting conservation areas.
Archaeology
Identifies hidden sites and maps historical ruins.
Provides insights into cultural landscapes and past civilization.
Passive Remote Sensing
In passive remote sensing, energy is naturally sourced, typically from the Sun. The system
records electromagnetic radiation (EMR) reflected or emitted by objects on the Earth's
surface. These systems generally operate using solar energy and work in specific
wavelengths that can traverse the atmosphere without significant loss. Passive sensors are
mounted on airborne or spaceborne platforms and capture radiation in the visible, infrared,
and thermal infrared regions.
Key Characteristics:
Dependent on daylight for energy, limiting data collection to daylight hours.
Susceptible to interference from weather conditions like cloud cover or rain.
Measures radiation reflected from objects or emitted based on their temperature.
Limitations:
Nighttime data collection is not possible.
Weather can hinder data collection, especially in visible and near-infrared bands.
Applications:
Vegetation monitoring: Health and coverage of crops and forests.
Land cover mapping: Identifying features such as water bodies, urban areas, and natural
landscapes.
Oceanography: Observing sea surface temperature, currents, and ocean color.
Examples:

5
Landsat Satellites: Long-term observation for environmental monitoring and land use.
MODIS: Used for global vegetation and atmospheric monitoring.
Active Remote Sensing
Active remote sensing generates its own energy, typically through microwaves, radar, or
lasers, which is then directed towards the target. The energy reflected or scattered by the
target is recorded by the sensor. This method is capable of functioning at night or in cloudy
conditions.
Key Characteristics:
Uses self-generated energy, unlike passive systems that rely on sunlight.
Can operate independently of weather and daylight.
Provides high-resolution data, particularly for topographic mapping and vegetation
structure.
Applications:
Radar (SAR): High-resolution imaging for mapping Earth's surface and monitoring urban
infrastructure.
LiDAR: Used for creating 3D models of terrain, forest structures, and buildings.
Disaster management: Assessing damage from events like earthquakes or floods.
Examples:
TerraSAR-X: Provides detailed radar imagery for Earth observation.
LiDAR systems: Used for precise measurements in forestry, coastal studies, and
environmental monitoring.

6
7
Advantages of Active Sensing
• Independence from Illumination: Operates effectively day and night.
• All-Weather Capability: Functions in adverse weather, including rain and fog.
• 3D Mapping: Provides detailed three-dimensional terrain and structural data.
• High Spatial Resolution: Captures precise and detailed features.
• Vegetation and Subsurface Analysis: Penetrates canopies or ground layers for
detailed studies.
• Dynamic Monitoring: Tracks real-time changes like floods or landslides.
• Advantages of Passive Sensing
• Cost-Effective: Utilizes natural light, reducing operational costs.
• Wide Area Coverage: Captures extensive regions in a single pass.
• High Spectral Resolution: Detects diverse wavelengths for detailed spectral analysis.
• Temporal Analysis: Monitors long-term changes with historical data.
• Non-Intrusive: No energy emission, ensuring environmental safety.
• Abundant Data Availability: Access to continuous data from established satellite
missions.

Sensor Platform
Definition
A sensor platform is a device or vehicle from which a sensor is mounted to collect information about
an object or target. The platform height, type, and design influence spatial resolution and the area
observed.
Types of Platforms
Sensor platforms are categorized based on altitude and design. They are mainly divided into
three types:
1. Ground-Based Platforms
2. Airborne Platforms
3. Spaceborne Platforms
[Link]-Based Platforms
Altitude Range: 0 meters (Earth’s surface).
Description: Used to record detailed information about objects or features on Earth's
surface. These platforms are stationary or mobile and provide highly accurate ground truth
data.
.
Examples:
• Handheld platforms: Used for small-scale measurements.
• Cherry pickers and towers: Used for mounting sensors at different heights for land
feature analysis.
• Vehicles: Used for mobile surveys and sensor data collection.
Applications:
• Portable cameras

8
 • Ground truth verification.
• Environmental and land feature studies.
• Sensor design testing.
• spectroradiometers are widely used for reference data and ground truth verification.
• Aid in scientific studies of signal-object and signal-sensor interactions.
• Employed in both laboratory and field studies for sensor design and land feature
characterization
2. Airborne Platforms
Altitude Range: Low-altitude (below 30,000 feet), High-altitude (30,000–50,000 feet).
Description: These platforms provide high-resolution data and are used for detailed
imagery collection, environmental monitoring, and disaster response.
Examples:
• Aircraft: Fixed-wing or rotary-wing aircraft for various surveys.
• Balloons: Used for collecting imagery at altitudes of around 30 km.
• Drones: Miniature, remotely piloted aircraft equipped with sensors for imagery,
infrared, radar, and TV surveillance. They operate between 100 meters and 5,000
meters (5 km).
• Rockets: High-altitude sounding rockets used for synoptic imagery and large-area
coverage.
Applications:
• Disaster response (e.g., fire or flood monitoring).
• Environmental surveys and detailed imagery collection.
• Local area monitoring and research.
Advantages:
• High spatial resolution for detailed imagery.
• Versatility in carrying different sensor types (e.g., cameras, infrared, radar).
• Cost-effective and flexible, especially drones for low-cost operations.
• Provides near-real-time data for emergency response and dynamic monitoring.
• Useful for large-area coverage in aerial surveys and environmental monitoring.

3. Spaceborne Platforms
Altitude Range: 250 km to 36,000 km.
Description: Sensors mounted on spacecraft orbiting Earth. These platforms are crucial
for large-area coverage, frequent observations, and collecting quantitative data over a
long period.
Types:
• Low Earth Orbit (LEO): 160 km to 2,000 km (e.g., Landsat, MODIS).
• Medium Earth Orbit (MEO): 2,000 km to 35,000 km (e.g., GPS satellites).
• Geostationary Orbit (GEO): Positioned at 36,000 km (e.g., weather satellites like
GOES).

9
 Applications:
• Climate studies, land-use change monitoring, weather forecasting.
• Environmental monitoring (e.g., deforestation, urbanization).
• Agriculture, disaster monitoring, and large-scale surveys.
Advantages:
• Large area coverage.
• Frequent, repetitive observations.
• Radiometrically calibrated sensors for accurate measurements.
• Cost-effective for global monitoring.
Examples:
• Landsat: Provides multispectral images for land-use analysis.
• SPOT: French satellite for high-resolution Earth observation.
• MODIS: Monitors global environmental changes.

Electromagnetic Spectrum
Remote sensing requires an energy source to illuminate the target unless the energy is
emitted by the target itself.
The energy used is in the form of electromagnetic radiation (EMR).
Electromagnetic Radiation (EMR) Basics
EMR consists of two perpendicular components:
Electric Field (E): Varies in magnitude and is perpendicular to the direction of radiation
travel.
Magnetic Field (M): Oriented at right angles to the electric field.
Both fields travel at the speed of light (c = 3 × 10⁸ m/s)
Properties of EMR Relevant to Remote Sensing
Wavelength (λ):
The length of one wave cycle, measured as the distance between successive wave crests.
Units of measurement:
Metres (m): Base unit.
Nanometres (nm): 10-9 metres.
Micrometres (μm): 10-6 metres.

10
Centimetres (cm): 10-2 metres.
Wavelength is denoted by the Greek letter λ (lambda).

Frequency (f):
The number of wave cycles passing a fixed point per second.
Measured in Hertz (Hz): One cycle per second.
Multiples: kHz (kilohertz), MHz (megahertz), GHz (gigahertz).

Relationship Between Wavelength and Frequency:

Shorter Wavelength → Higher Frequency.

Longer Wavelength → Lower Frequency.
These properties help determine:
The type of EMR interacting with the target.

11
The kind of information extractable from remote sensing data.
Electromagnetic Spectrum Overview
The EM spectrum covers a wide range of wavelengths and frequencies.
Ranges from shorter wavelengths (high energy) to longer wavelengths (low energy):
Shorter Wavelengths: Gamma rays, X-rays.
Medium Wavelengths: Visible light, infrared.
Longer Wavelengths: Microwaves, radio waves.
Useful Regions of the Electromagnetic Spectrum for Remote Sensing
Gamma Rays and X-rays:
Shortest wavelengths, highest energy.
Limited application in remote sensing due to atmospheric absorption.
Ultraviolet (UV):
Short wavelengths just beyond visible light.
Useful for specific atmospheric studies.
Visible Light:
Wavelength range: 0.4–0.7 μm.
Perceived by the human eye, commonly used in optical remote sensing.
Infrared (IR):
Divided into near-infrared (NIR), mid-infrared (MIR), and thermal infrared (TIR).
Applications:
Vegetation and water studies (NIR and MIR).
Temperature and heat mapping (TIR).
Microwaves:
Longer wavelengths (mm to cm range).
Can penetrate clouds, rain, and vegetation.
Useful for radar and all-weather remote sensing.
Radio Waves:
Longest wavelengths, lowest energy.
Applications in communication and broadcasting.

Theoretical Models in Remote Sensing

Models are developed to understand EMR interactions with different surfaces and media:
Anisotropic Effects: Direction-dependent properties.
Random Distribution of Scatters: Interaction with distributed particles.
Rough Surface Effects: Interaction with uneven surfaces.
Light as an Electromagnetic Wave
Light can be considered as a wave propagating through the electromagnetic field of the
universe.
Understanding light’s wave properties helps in interpreting remote sensing data.

12
13
14
+

15
16
Radiation Laws and Related Concepts

17
18
19
Atmospheric Interactions with electromagnetic
radiations:
All electromagnetic radiation detected by a remote sensor must pass
through the atmosphere twice—before and after interacting with
Earth's surface. This passage significantly influences the
characteristics of the radiation, such as speed, frequency, intensity,
spectral distribution, and direction. The primary processes involved
are scattering and absorption.
1. Scattering in the Atmosphere
Definition
Scattering is the process by which electromagnetic radiation (EMR)
interacts with gases and particles in the atmosphere, deviating from
its straight-line path. This interaction can change the direction and
intensity of the radiation, and it depends on factors such as the size
of the particles, the wavelength of the radiation, and the
atmospheric composition.

Types of Scattering
20
1. Selective Scattering:
This type of scattering depends on the wavelength of the
radiation and the size of the particles causing the scattering.
o Rayleigh Scattering:

• Occurs when the particles (e.g., air molecules) are much

smaller than the wavelength of the radiation.
• Shorter wavelengths (e.g., blue and violet light) are scattered
more than longer wavelengths.
• Example: The blue appearance of the sky during the day is due
to Rayleigh scattering, as blue light is scattered more than red
light.
• Impact on Remote Sensing: Rayleigh scattering can reduce the
clarity of satellite images, especially in the visible spectrum.
o Mie Scattering:
• Caused by particles similar in size to the wavelength of
radiation, such as dust, smoke, and water droplets.
• All wavelengths are scattered more evenly than in Rayleigh
scattering, but longer wavelengths (e.g., red light) are affected
more.
• Example: The hazy appearance of the sky on smoggy or misty
days is a result of Mie scattering.
• Impact on Remote Sensing: Reduces image contrast, especially
in polluted or foggy conditions.
o Raman Scattering:
• A type of selective scattering where light interacts with
molecules, causing a change in wavelength.
• Results in slight spectral shifts, though its impact is less
significant in remote sensing compared to Rayleigh and Mie
scattering.
• Example: Used in spectroscopy for studying molecular
compositions but has limited influence on satellite imagery.
2. Non-Selective Scattering:

21
 • Occurs when the particles are much larger than the
wavelength of the radiation, such as water droplets and ice
particles in clouds.
• All wavelengths are scattered equally, making objects appear
white or washed out.
• Example: The white color of clouds results from non-selective
scattering of visible light by large water droplets.
• Impact on Remote Sensing: Obstructs surface details in
satellite imagery under cloudy conditions.

Impact of Scattering on Remote Sensing

1. Image Clarity: Scattering reduces the sharpness and contrast of satellite
images, especially in the visible and near-infrared bands.
2. Radiometric Errors: Scattering contributes to errors in radiance
measurements, requiring atmospheric correction techniques.
3. Wavelength Selection: Remote sensing systems are designed to use
wavelengths less affected by scattering, such as the near-infrared and
microwave regions, to improve data accuracy.

22
ATMOSPHERIC PROPERTIES
• The atmosphere and the Sun interact to determine the radiation
measured by satellites or aircraft. Understanding the atmosphere’s
properties and how solar radiation behaves is essential for accurate
remote sensing.
1. The Role of Atmosphere in Radiation
• Most of the radiance measured by high-altitude aircraft or satellites
originates from multiple scattering in the atmosphere.
• To accurately measure signals from suspended particles (like aerosols or
clouds), scientists must remove the atmospheric contribution through
careful correction techniques.

23
2. Solar Radiation as the Primary Source
• The Sun emits radiation across a wide range of wavelengths, known as
shortwave radiation.
• At the top of the Earth’s atmosphere, this incoming radiation is called the
extraterrestrial irradiance or the solar constant.

3. Key Factors Affecting Solar Radiation

• Sun’s Photosphere:
o Acts like a black body, emitting radiation in a broad spectrum.
o Determines the overall intensity of solar radiation.
• Sun’s Chromosphere:
o Plays a role in scattering and absorption processes, altering the
solar radiation.
o Contains Fraunhofer lines, which are prominent drops in the
spectral distribution due to strong absorption by elements in the
chromosphere.

4. Figure Explanation (Fig. 2.11)

• The graph shows the solar irradiance at the top of the Earth’s
atmosphere over the wavelength range 0.4–0.8 μm.
• Observations:
o The irradiance is highest at around 1,200 W/m²/μm near 0.5 μm
and decreases as wavelength increases.
24
 o Dips in the curve represent Fraunhofer lines, caused by
absorption processes in the Sun’s chromosphere.

Why Is This Important?

• Atmospheric Corrections: Before analyzing satellite data, scientists must
account for atmospheric effects like scattering and absorption.
• Radiative Transfer Calculations: Models require detailed knowledge of
atmospheric parameters to simulate how radiation behaves.
• Remote Sensing Accuracy: Understanding solar radiation and
atmospheric interaction ensures precise environmental monitoring.

OZONE ABSORPTION
Ozone is a crucial trace gas in the Earth's atmosphere, primarily found in the
stratosphere, where it plays a key role in absorbing harmful ultraviolet (UV)
radiation. It helps regulate the Earth's radiation balance by filtering out high-
energy solar radiation.

1. Ozone Absorption in the Atmosphere:

• Ozone exists primarily in the stratosphere between 20-40 km altitude,
with the highest concentration around 25 km.
• At these heights, ozone absorbs shortwave radiation (solar radiation),
especially ultraviolet (UV) radiation. This absorption helps in regulating
the Earth's energy balance.
• Ozone has the most significant effect on radiation at these heights, and
its role diminishes at lower altitudes.

2. Chappuis Band of Ozone:

• Ozone in the visible spectrum absorbs light in the Chappuis band, a
specific wavelength region.
• This absorption is important for remote sensing, especially in detecting
oceanic constituents from space.

3. Absorption of Chlorophyll Fluorescence:

25
 • Chlorophyll in plants emits fluorescence (light) when exposed to
sunlight. However, this fluorescence is affected by water vapor (H₂O)
and oxygen (O₂) in the atmosphere.
• Water vapor and oxygen absorb the fluorescence in their vibration
action bands, which prevents some of the emitted light from reaching
the top of the atmosphere.

4. Radiative Transfer and Transmission Functions:

• Radiative transfer refers to how sunlight interacts with the atmosphere,
including how it's absorbed or scattered by gases and particles.
• Transmission functions for oxygen (O₂) and water vapor (H₂O) are
calculated to understand how much sunlight is absorbed or scattered by
these gases.
• The absorption is modeled using Lorentz's theory of collision
broadening, which describes how molecules' absorption lines get
broader due to molecular collisions.
• Doppler broadening (caused by the movement of molecules) and
resonance broadening are neglected in this case because they have
minimal impact.

5. Wavelength Interval Averaging:

• The transmission functions are averaged over small intervals (typically 1
nm) to simplify calculations and to account for the variation in
absorption and scattering across different wavelengths.
• Solar flux reduction occurs due to the absorption and scattering by the
atmosphere. For instance, in mid-latitude summer, clear skies cause a
reduction in the solar energy reaching Earth’s surface.

6. Temperature and Pressure Effects:

• Studies show that temperature has a minimal effect on the transmission
of solar radiation through the atmosphere.
• Pressure plays a significant role in transmission and absorption
processes, and it must be considered in radiative transfer models.

7. Molecular Scattering and Rayleigh Scattering:

26
 • Molecular scattering happens when air molecules (like nitrogen and
oxygen) interact with sunlight.
• This scattering follows Rayleigh scattering theory, where small
molecules scatter light primarily based on the wavelength of the light.
• Pressure influences the scattering, and the scattering coefficient (which
tells us how much light gets scattered) can be estimated from
atmospheric pressure data.

8. Aerosol Scattering:
• Aerosols (tiny particles in the atmosphere, like dust or smoke) also
scatter sunlight and cause atmospheric turbidity (cloudiness).
• Aerosols affect the transmission of sunlight through the atmosphere.
Their scattering properties depend on factors like concentration,
composition, and size distribution.
• The vertical distribution of aerosols (how they are spread throughout
the atmosphere) is studied and helps estimate their effect on scattering.
• Aerosol scattering is modeled using Mie theory, which describes how
particles scatter light in the visible and near-infrared spectrum.
• The turbidity or cloudiness of the atmosphere is measured in terms of
horizontal visibility, which can range between 6 km to 88 km. Aerosol
scattering becomes stronger when there are more particles in the
atmosphere.

Summary of How It All Works:

1. Ozone in the stratosphere absorbs harmful UV radiation, which helps
protect the Earth from excess energy absorption.
2. The Chappuis band in the visible spectrum of ozone is crucial for
detecting oceanic features from space.
3. Water vapor and oxygen absorb chlorophyll fluorescence, limiting how
much light reaches the atmosphere.
4. Radiative transfer models are used to understand how sunlight interacts
with the atmosphere, factoring in transmission functions for gases like
O₂ and H₂O.
5. Temperature has a minor effect on transmission, while pressure
significantly impacts how sunlight is absorbed or scattered.

27
 6. Rayleigh scattering by air molecules and aerosol scattering by tiny
particles both affect how sunlight is scattered in the atmosphere.
7. These scattering effects are modeled using Mie theory for aerosols and
Rayleigh theory for molecules.
In conclusion, sunlight interacts with various atmospheric components (ozone,
gases, molecules, and aerosols) through absorption and scattering, affecting
the amount of energy that reaches Earth's surface. Understanding these
processes is key for analyzing environmental data from remote sensing.

Atmospheric Effects on Spectral Response Patterns

The energy captured by a sensor is significantly influenced by the
atmosphere between the sensor and the Earth's surface. This interaction
alters the spectral response patterns, which are essential for remote
sensing applications.
There are two main atmospheric effects on the radiance recorded by the
sensor:
1. Attenuation (Reduction in Energy): The atmosphere reduces or
attenuates the energy that is transmitted to the ground object,
meaning that the amount of energy reaching the object is lower
than the original incident energy.
2. Atmospheric Reflection (Path Radiance Addition): The atmosphere
itself acts as a reflector, adding path radiance (scattered energy) to
the signal detected by the sensor, which can result in additional
unwanted radiation being recorded.
These two effects can be mathematically described as follows (Lillesand
and Kiefer, 1979):
• Attenuation: The reduction in energy is due to absorption and

scattering by the atmospheric gases and aerosols.

• Path Radiance: The scattered energy from the atmosphere adds to

the signal received by the sensor, affecting the accuracy of

measurements.
Together, these effects modify the radiance values, which can impact the
analysis of ground features in remote sensing data.

28
29
30
31
Spectral Reflectance Curves
Spectral reflectance curves show how different materials (like water,
soil, or vegetation) reflect sunlight or other radiation at different
wavelengths. These curves help us "see" things that our eyes can't,
allowing remote sensing systems to identify and analyze Earth
features.
It is defined as the wavelength reflected to object to the energy
of wavelength incident on the object, such reflectance is called
spectral reflectance

ρλ = ER(λ)
EI(λ)
ρλ : Reflantance of object (%)
ER(λ) : Energy of wavelength reflected from the object
EI(λ) : Energy of wavelength incident on object
λ : wavelength(μm)
What is Spectral Reflectance?
• When sunlight hits a surface, some of it is absorbed, some is

reflected, and some passes through (transmitted).

• Spectral reflectance measures how much light a surface reflects

at specific wavelengths (colors of light).

• Each material has a unique "reflection fingerprint" called its

spectral signature.

Factors Influencing Reflectance:

32
• Solar Orientation:
o The sun's position, defined by the solar azimuth angle

and solar elevation angle, determines the intensity and

angle of incident radiation.
• Sensor Orientation:
o The angle at which the sensor views the target (nadir or

oblique view) impacts the recorded reflectance.

✓ Every object on the surface of the earth has its unique spectral
reflectance, below image shows the avg spectral reflectance
curves for three typical earth’s features: Vegetation, soil and
water

33
o Spectral reflectance curves plot reflectance (yyy-axis)
against wavelength (xxx-axis).
o The shape of the curve is unique for different materials
(e.g., vegetation, soil, water) and serves as their spectral
signature.
o Vegetation:
▪ Low reflectance in blue (400-500 nm) and red (600-

700 nm) due to chlorophyll absorption.

▪ High reflectance in green (500-600 nm), which

makes vegetation appear green to our eyes.

▪ Very high reflectance in near-infrared (700-1300 nm)

due to the leaf's cellular structure.

▪ Absorbs most blue and red light (for
photosynthesis).
▪ Reflects green light (why plants look green).
▪ Reflects even more in near-infrared (used to
measure plant health).

34
 o Water:
▪ Low reflectance in all visible wavelengths; appears

dark.
▪ Strong absorption in infrared (>700 nm), which

makes water easily distinguishable in satellite

images.
▪ Reflects very little light, so it looks dark in satellite
images.
▪ Absorbs most light, especially in infrared, which
helps detect water bodies.
o Soil:
▪ Reflectance varies with moisture, organic content,

and texture.
▪ Generally increases with wavelength in the visible

and near-infrared regions.

▪ Reflectance depends on factors like moisture and
organic content.
▪ Dry soil reflects more light, while wet soil absorbs
more.

Factors Influencing Spectral Reflectance Curves

1. Material Properties:
o Composition, texture, and structure affect how materials

absorb and reflect radiation.

2. External Conditions:
o Moisture levels, sunlight angle, and time of day impact

reflectance.
o For example, wet soil reflects less light than dry soil.

3. Atmospheric Effects:
o Scattering and absorption by gases, aerosols, and water

vapor alter the observed reflectance values.

35
Curve Features and Analysis
1. Peaks and Valleys:
o Peaks indicate high reflectance at certain wavelengths.

o Valleys represent absorption bands specific to the

material.
o Example: Vegetation has a prominent "red-edge" between

red and near-infrared due to a sharp increase in

reflectance.
2. Multi-Spectral and Hyperspectral Data:
o Multi-spectral sensors capture reflectance at a few

discrete bands.
o Hyperspectral sensors provide detailed reflectance curves

across hundreds of narrow bands, enabling precise

material differentiation.

Applications of Spectral Reflectance Curves

1. Vegetation Health Monitoring:

o Healthy plants have high near-infrared reflectance and

low red reflectance (e.g., using NDVI).

2. Water Quality Analysis:
o Polluted water reflects differently compared to clean

water, especially in the visible and infrared bands.

3. Soil Studies:
o Helps in analyzing soil moisture, mineral content, and

organic matter.

36
Satellite
A satellite is a celestial body or man-made object that orbits around a larger
body, such as a planet or a star. Natural satellites include the Moon (Earth’s
natural satellite), while artificial satellites are human-engineered devices
launched into space to perform specific tasks. Satellites rely on the principles
of gravity and inertia to remain in orbit.

Classification of Satellites
Satellites are classified based on
• Function,
• Orbit, and
• Application:

1. Based on Function
• Communication Satellites: Enable telecommunication services like TV
broadcasts, internet, and phone networks (e.g., INSAT series).
• Navigation Satellites: Provide positioning and timing information (e.g.,
GPS, GAGAN).
• Earth Observation Satellites: Monitor weather, natural resources, and
environmental changes (e.g., IRS, Landsat).
• Scientific Satellites: Conduct space research and study celestial
phenomena (e.g., Hubble Space Telescope).
• Military Satellites: Used for surveillance, reconnaissance, and defense
purposes.

2. Based on Orbit
• Geostationary Satellites (GEO):
o Remain fixed relative to a point on Earth.
o Operate at an altitude of ~36,000 km.
o Used for weather forecasting and communication.
• Low Earth Orbit Satellites (LEO):
o Positioned at altitudes between 200 km to 2,000 km.
o Used for Earth observation and space exploration.
• Medium Earth Orbit Satellites (MEO):
o Altitudes between 2,000 km to 35,000 km.

37
 o Mainly used for navigation systems like GPS.
• Polar Orbiting Satellites:
o Traverse over Earth's poles, providing global coverage.
o Ideal for climate monitoring and mapping.

3. Based on Application
• Weather Satellites: Predict and monitor climatic changes (e.g.,
METEOSAT).
• Spy Satellites: Perform military surveillance.
• Astronomical Satellites: Study space and stars (e.g., James Webb Space
Telescope).
• Commercial Satellites: Used for TV broadcasting, internet, and business
applications.

Satellite orbits
Satellite orbits are specific paths that satellites follow as they revolve around
the Earth or other celestial bodies. These orbits are governed by the
gravitational pull of the Earth and the satellite's velocity. Depending on their
purpose, satellites are placed in different types of orbits, each with unique
characteristics.
Types:
satellite orbits are broadly classified based on:
▪ Altitude
▪ Inclination
▪ Eccentricity

38
A. Based on Altitude:
1. Low Earth Orbit (LEO):
o Altitude: 180 km to 2,000 km above Earth.
o Features:
▪ Short orbital periods (90-120 minutes).
▪ High resolution for imaging applications.
▪ Frequently used for Earth observation, weather monitoring,
and communication (e.g., Starlink satellites).
o Applications:
▪ Earth observation (e.g., Landsat, Sentinel).

▪ International Space Station (ISS) is also in LEO.

▪ Earth Observation: High-resolution imaging for agriculture,
forestry, urban planning, and resource mapping.
▪ Scientific Research: Studying Earth’s atmosphere, magnetic
field, and environmental changes.
▪ Communication: Satellite constellations for low-latency internet
(e.g., Starlink, OneWeb).
▪ Space Exploration: Hosting space stations like the International
Space Station (ISS).
2. Medium Earth Orbit (MEO):
o Altitude: 2,000 km to 35,786 km above Earth.
o Features:
▪ Longer orbital periods (2 to 12 hours).
▪ Used for navigation systems.
o Applications:
▪ GPS, GLONASS, and Galileo navigation satellites.
▪ Navigation Systems: GPS, GLONASS, Galileo, and BeiDou for
global positioning.
▪ Weather and Environmental Studies: Monitoring global
weather patterns and sea-level changes.
▪ Communication: Regional data relay and voice services.
3. Geostationary Orbit (GEO):
o Altitude: Exactly 35,786 km above Earth’s equator.
o Features:
▪ Orbital period matches Earth’s rotation (24 hours).

39
 ▪ Appears stationary relative to a fixed point on Earth.
▪ Covers one-third of the Earth’s surface.
o Applications:
▪ Weather satellites (e.g., INSAT, METEOSAT).
▪ Communication satellites for broadcasting and
telecommunications.
▪ Weather Monitoring: Real-time weather updates and
forecasting (e.g., INSAT, METEOSAT).
▪ Telecommunications: Satellite TV, radio, and internet
broadcasting.
▪ Navigation: Supporting ground-based navigation systems.
▪ Disaster Management: Monitoring cyclones, hurricanes, and
environmental hazards.
4. High Earth Orbit (HEO):
o Altitude: Above 35,786 km.
o Features:
▪ Long orbital periods.
▪ Used for specific scientific missions.
o Applications:
▪ Monitoring space weather, deep-space missions.
▪ High-Latitude Communication: Ensures coverage for polar
regions (e.g., Molniya orbits).
▪ Surveillance: Monitoring high-latitude regions for military or
scientific purposes.
▪ Astronomical Observations: Capturing unique celestial events
from specific vantage points.
B. Based on Inclination:
1. Equatorial Orbit:
o Satellite revolves around the Earth along the equator.
o Inclination: 0°.
o Example: Geostationary satellites.
2. Polar Orbit:
o Satellite passes over the Earth’s poles during each revolution.
o Inclination: Close to 90°.

40
 o Covers the entire Earth's surface over time due to the planet’s
rotation.
o Example: Earth observation and reconnaissance satellites.
o Applications
o Global Monitoring: Imaging the entire Earth surface over time for
mapping and resource assessment.
o Climate Studies: Monitoring glaciers, polar ice caps, and
environmental changes.
o Reconnaissance: Surveillance and defense applications for military
purposes.
3. Sun-Synchronous Orbit:
o A type of polar orbit that maintains a constant position relative to
the Sun.
o Allows satellites to capture images with consistent lighting
conditions.
o Example: SPOT, Landsat satellites.
o Applications:
o Environmental Monitoring: Consistent lighting conditions for
imaging (e.g., vegetation health, water quality).
o Mapping: Producing accurate and comparable maps over time.
o Remote Sensing: High-resolution imagery for disaster
management and urban planning.

C. Based on Eccentricity:
1. Circular Orbit:
o The distance between the satellite and Earth remains constant.
o Example: GPS satellites.
2. Elliptical Orbit:
o The satellite moves closer to Earth at the perigee and farther away
at the apogee.
o Useful for covering regions like high-latitude areas.
o Example: Molniya orbit (used by Russian satellites).

Key Parameters of Satellite Orbits

41
 • Altitude: Distance from the Earth’s surface.
• Inclination: Angle of the satellite's orbit relative to the equatorial plane.
• Period: Time taken for one complete revolution.
• Eccentricity: Determines whether the orbit is circular or elliptical.
• Semi-Major Axis: Average distance of the satellite from Earth.
• Apogee & Perigee: Farthest and closest points of the satellite from Earth
in an elliptical orbit.

Real-Life Example

The International Space Station (ISS) orbits in LEO, traveling at a speed of 28,000 km/h and
completing 16 orbits of the Earth per day. It provides a platform for scientific research and
international collaboration.

Satellite Orbits (overview 2M OR 3M)

1. Geostationary Orbit (GEO)
o Fixed position relative to Earth.
o Used for weather monitoring, telecommunications, and
broadcasting (e.g., TV satellites).
2. Geosynchronous Orbit
o Similar to GEO but slightly inclined; not fixed above the equator.
o Applications include communications and navigation.
3. Polar Orbit
o Passes over Earth's poles; provides global coverage.
o Used for Earth observation, weather monitoring, and
reconnaissance.
4. Sun-Synchronous Orbit (SSO)
o Maintains consistent lighting for imaging at the same time each
day.
o Used for environmental monitoring, mapping, and remote
sensing.
5. Low Earth Orbit (LEO)
o Close to Earth's surface (up to 2,000 km altitude).
o Applications: Earth observation, communication (e.g., Starlink),
and ISS operations.
6. Medium Earth Orbit (MEO)
o Between LEO and GEO (2,000-35,000 km altitude).
o Used for navigation systems like GPS, GLONASS, and Galileo.
7. Highly Elliptical Orbit (HEO)

42
 o Elliptical shape; closer to Earth at one end (perigee) and far at the
other (apogee).
o Applications: Communications in polar regions and scientific
studies.
8. Inclined Orbit
o Not aligned with the equator; inclined relative to Earth's
equatorial plane.
o Applications: Surveillance and imaging.
9. Transfer Orbit
o Temporary orbit used to transfer satellites to higher orbits.
o Example: Geostationary Transfer Orbit (GTO).
[Link] Orbit
o A type of HEO with a long dwell time over a specific region.
o Used for communications and weather monitoring in high-latitude
areas.
[Link] Orbit
o Aligns with Earth's equator.
o Applications: Communication and environmental observation in
equatorial regions.
[Link] Orbit
o Moves in the opposite direction of Earth's rotation.
o Applications: Military and reconnaissance.
[Link]-Earth Lagrange Point Orbits
o Position satellites in stable points between the Earth and Sun.
o Applications: Space weather monitoring (e.g., SOHO) and deep
space observations (e.g., James Webb Telescope).
[Link] Orbit
o Orbits between Earth and the Moon.
o Applications: Lunar exploration and communications for Moon
missions.
[Link] Orbit
o Orbits around libration points (gravitational balance points).
o Applications: Space exploration (e.g., Gateway station near the
Moon).

43
Perigee and Apogee are the terms related to orbit of an object around earth
(or another celestial body)
1. perigee: The point in an object’s orbit that is closest to Earth
Eg : when the moon is at perigee, it appears slightly larger and brighter in
the sky.
2. Apogee: The point in an object that is farthest from the earth.
Eg : when the moon is at apogee, it appears slightly smaller and dimmer in
the sky.
44
Sensor Resolution
Sensor resolution refers to the ability of a remote sensing sensor to
detect and distinguish fine details of the Earth's surface.
• It is classified into four types, each critical for interpreting
satellite data accurately:
1. Spatial Resolution.
2. Spectral Resolution.
3. Temporal Resolution.
4. Radiometric Resolution.
Spatial Resolution
Definition
Spatial resolution refers to the smallest area on the ground that a single pixel in
an image represents. It determines the level of detail that a sensor can capture.
For example:
• A 10m spatial resolution means each pixel represents a 10m × 10m area.
• A 1m resolution offers much finer detail than a 30m resolution.
Explanation : Spatial resolution is how much detail you can see in an image.
Imagine looking at a map or satellite image – each small dot (pixel) on the
image represents a specific area on the ground. The smaller the area each pixel
covers, the higher the resolution and the clearer the details you can see. So,
high spatial resolution means more detail, and low resolution means less detail.

Importance of Spatial Resolution

1. Determines Clarity: Higher resolution allows for clearer and more
detailed imagery.
2. Application-Specific:
o High resolution is critical for detailed applications like urban
planning and infrastructure mapping.

45
 o Low resolution is suitable for large-scale studies like global climate
or ocean monitoring.

Types of Spatial Resolution : High, Medium, low

1. High Spatial Resolution
o Definition: Resolves features smaller than 5 meters.
o Examples:
▪ WorldView-3: 0.31m resolution.
▪ Cartosat-2: 0.8m resolution.
o Applications:
▪ Urban mapping and detailed infrastructure planning.
▪ Precision agriculture for detecting small-scale changes in
crops.
2. Medium Spatial Resolution
o Definition: Resolves features between 5m and 30m.
o Examples:
▪ Sentinel-2: 10m resolution for visible and NIR bands.
▪ Landsat-8: 30m resolution for multispectral imaging.
o Applications:
▪ Regional land-use and land-cover mapping.
▪ Monitoring deforestation and forest health.
3. Low Spatial Resolution
o Definition: Resolves features larger than 250 meters.
o Examples:
▪ MODIS: 250m resolution for red and NIR bands.
▪ NOAA AVHRR: 1km resolution.
o Applications:

46
 ▪ Global climate monitoring.
▪ Ocean and atmospheric studies.

Factors Affecting Spatial Resolution

1. Altitude of the Sensor:
o Higher the altitude, the larger the area covered by a pixel,
resulting in lower spatial resolution.
o Example: Satellites in Low Earth Orbit (LEO) have higher resolution
than those in Geostationary Orbit (GEO).
2. Sensor Design:
o Advanced sensors, such as panchromatic sensors, often have
higher resolution than multispectral ones.
3. Ground Sampling Distance (GSD):
o The physical distance between the centers of adjacent pixels on
the ground.
o Smaller GSD means higher resolution and finer details.
4. Swath Width:
o Wider swath widths (area covered by a sensor in one pass) usually
reduce spatial resolution.

Applications of Spatial Resolution

1. Urban Planning:
o High-resolution imagery helps map roads, buildings, and other
urban infrastructure.
o Example: WorldView-3 for city-level planning.
2. Agriculture:
o Medium-resolution sensors monitor crop health, detect irrigation
patterns, and map soil quality.

47
 o Example: Sentinel-2 for vegetation monitoring.
3. Climate Studies:
o Low-resolution sensors track global phenomena such as sea-
surface temperatures, snow cover, and vegetation index.
o Example: MODIS for global environmental changes.
4. Disaster Management:
o High-resolution data is crucial for mapping disaster-affected areas,
such as flooded zones or earthquake damage.
5. Military and Surveillance:
o High spatial resolution is used for reconnaissance and identifying
small-scale features like vehicles or structures.

Advantages of High Spatial Resolution

• Enables detailed analysis of small-scale features.
• Facilitates accurate mapping and infrastructure planning.
• Useful for applications like precision farming and disaster response.

Limitations of Spatial Resolution

1. High Cost: High-resolution sensors and data are more expensive to
develop and access.
2. Data Size: Produces large datasets that require advanced storage and
processing systems.
3. Reduced Coverage: Narrow swath widths limit the area covered per
pass.
4. Atmospheric Effects: High-resolution imagery is more sensitive to
distortions caused by atmospheric interference.

Conclusion

48
Spatial resolution is a critical parameter in remote sensing that directly
influences the clarity and usability of captured images. High-resolution sensors
are ideal for applications requiring detailed analysis, while medium and low
resolutions balance detail with broader coverage. Proper selection of spatial
resolution is essential to meet the objectives of specific studies or projects.

Spectral Resolution
Definition
Spectral resolution refers to the ability of a sensor to distinguish between
different wavelengths of light (or electromagnetic radiation) in the
electromagnetic spectrum. It determines how finely a sensor can divide the
spectrum into individual bands, allowing it to capture specific features based
on their spectral characteristics.
For example:
• A sensor with a spectral resolution of 10 bands can capture 10 distinct
wavelength ranges of light.
• A sensor with a higher spectral resolution can distinguish more subtle
differences between different wavelengths.
• Explanation : Spectral resolution is the ability of a sensor to detect and
distinguish different wavelengths (or colors) of light. Light, or
electromagnetic radiation, is made up of a range of wavelengths that
vary from visible light to infrared and beyond. Each material or object
reflects or absorbs different wavelengths in unique ways, creating a
"spectral signature."
• A sensor with high spectral resolution can capture many narrow bands of
wavelengths, enabling it to detect subtle differences between materials
based on their unique spectral characteristics. In contrast, a sensor with
low spectral resolution captures fewer, broader bands, which may miss
these subtle differences.
• For example, a high-resolution sensor might detect a specific plant’s
stress by identifying changes in narrow wavelength bands in the infrared

49
 range, whereas a lower resolution sensor might not capture these small
changes.
• Thus, spectral resolution helps sensors understand and analyze features
based on how they interact with different wavelengths of light, enabling
more precise identification and classification of materials and
phenomena

Importance of Spectral Resolution

• Increased Sensitivity: Higher spectral resolution allows sensors to detect
finer spectral differences, leading to better identification of materials and
features.
• Application-Specific:
o High spectral resolution is critical for detailed material
identification, such as distinguishing between different types of
vegetation or minerals.
o Lower spectral resolution is suitable for broader applications like
large-scale land classification and weather monitoring.
Types of Spectral Resolution
• High Spectral Resolution
o Definition: Distinguishes narrow and precise wavelength bands
(often in the range of 1 nm to 10 nm).
o Examples:
▪ Hyperion (NASA): 220 bands in the 400-2500 nm range.
▪ AVIRIS (Airborne Visible/Infrared Imaging Spectrometer):
High-resolution hyperspectral imagery.
o Applications:
▪ Mineral mapping and identification of specific materials.
▪ Vegetation classification based on distinct spectral features.
▪ Water quality monitoring using specific absorption features
in the spectrum.

50
 • Medium Spectral Resolution
o Definition: Captures a limited number of broader spectral bands
(around 10-20 bands).
o Examples:
▪ Landsat-8: 11 spectral bands including visible, near-infrared,
and shortwave infrared.
▪ Sentinel-2: 13 bands, covering from visible to near-infrared
wavelengths.
o Applications:
▪ Land-use/land-cover classification.
▪ Agricultural monitoring, including crop health analysis.
▪ Forest and vegetation studies.
• Low Spectral Resolution
o Definition: Captures only a few broad spectral bands (e.g., 2-4
bands).
o Examples:
▪ MODIS: 36 spectral bands for global environmental
monitoring.
▪ NOAA AVHRR: 5-6 broad bands for weather and climate
applications.
o Applications:
▪ Large-scale environmental monitoring (e.g., sea surface
temperature, cloud cover).
▪ Weather forecasting and atmospheric studies.
▪ Broad land classification.
Factors Affecting Spectral Resolution
• Sensor Design:
The ability to separate specific wavelength bands depends on the

51
 sensor’s optical system and detector technology. Hyperspectral sensors
capture finer spectral details than multispectral sensors.
• Bandwidth:
The width of each spectral band determines the level of spectral
resolution. Narrower bands allow finer spectral discrimination.
• Sampling Interval:
The more frequently a sensor samples wavelengths, the higher its
spectral resolution. For example, a sensor sampling every 10 nm will
have higher spectral resolution than one sampling every 100 nm.
Applications of Spectral Resolution
• Agriculture:
High spectral resolution helps differentiate between plant species, assess
crop health, detect pests, and monitor nutrient levels.
Example: AVIRIS for detecting plant stress based on specific spectral
signatures.
• Environmental Monitoring:
Spectral resolution helps track pollution levels, deforestation, and
changes in water bodies.
Example: MODIS for monitoring global climate phenomena.
• Mining and Geology:
High spectral resolution allows for the identification of minerals and
rocks based on their unique spectral signatures.
Example: Hyperion for mineral exploration.
• Urban Planning:
Spectral resolution helps differentiate between land types, building
materials, and vegetation.
Example: Landsat for urban growth and infrastructure planning.
Advantages of High Spectral Resolution
• Enables precise material identification based on spectral features.
• Ideal for detailed environmental monitoring and resource management.
• Supports complex analysis, such as detecting subtle changes in
vegetation health or mineral composition.

52
Limitations of Spectral Resolution
• Higher Cost: More advanced sensors with higher spectral resolution are
expensive to develop and operate.
• Large Data Volume: More bands result in larger datasets, requiring
significant storage and processing power.
• Complexity: Analyzing hyperspectral data requires advanced processing
techniques and expertise.
Conclusion
Spectral resolution is crucial in remote sensing, directly influencing the ability
to differentiate materials and monitor changes in the environment. High
spectral resolution is ideal for applications requiring detailed spectral
information, such as material identification, while medium and low spectral
resolution sensors balance data complexity and broader coverage. Proper
selection of spectral resolution is key to the success of specific applications.

Temporal Resolution
Definition
Temporal resolution refers to how often a sensor or satellite can capture data
for the same location over time. It determines the frequency with which an
image or observation of a specific area can be repeated. High temporal
resolution means more frequent data capture, while low temporal resolution
means less frequent capture.
For example:
• A satellite with a temporal resolution of one day captures an image of
the same area every day.
• A sensor with a temporal resolution of one month captures an image
only once a month.
Importance of Temporal Resolution
• Tracking Changes: Higher temporal resolution is essential for monitoring
fast-changing phenomena like weather patterns, vegetation growth, and
urban development.
• Application-Specific:
53
 o High temporal resolution is needed for real-time monitoring, such
as tracking natural disasters or crop health.
o Low temporal resolution is suitable for long-term studies like
climate change or large-scale mapping.
Types of Temporal Resolution
• High Temporal Resolution
o Definition: Captures data frequently, often multiple times a day or
every few hours.
o Examples:
▪ MODIS: Data collected every 1-2 days globally.
▪ Sentinel-1: Provides data every 6-12 days.
o Applications:
▪ Disaster management (e.g., tracking floods, wildfires, or
hurricanes).
▪ Monitoring agriculture (e.g., checking crop growth or soil
moisture).
▪ Weather forecasting and atmospheric analysis.
• Medium Temporal Resolution
o Definition: Data captured every few days to weeks.
o Examples:
▪ Landsat-8: Data captured approximately every 16 days.
▪ Sentinel-2: Data captured every 5 days (with two satellites).
o Applications:
▪ Land cover changes (e.g., urbanization, deforestation).
▪ Monitoring vegetation health over time.
▪ Studying changes in water bodies or lakes.
• Low Temporal Resolution

54
 o Definition: Data captured infrequently, such as once a month or
longer.
o Examples:
▪ Landsat-5: Data captured every 16 days but less frequently
in certain years.
▪ ASTER: Data captured once every 16-18 days.
o Applications:
▪ Long-term environmental monitoring (e.g., global climate
change).
▪ General land classification and mapping.
▪ Monitoring seasonal changes in vegetation.
Factors Affecting Temporal Resolution
• Orbital Characteristics:
The orbit of a satellite or sensor affects how frequently it can revisit the
same area. Low Earth Orbit (LEO) satellites usually have higher temporal
resolution, while satellites in higher orbits, like Geostationary Orbit
(GEO), have lower temporal resolution.
• Sensor Capacity:
Some sensors can capture data over large areas quickly, allowing for
higher temporal resolution, while others might take longer to scan the
same area, limiting their temporal resolution.
• Swath Width:
A wider swath width (the area the sensor captures at once) can reduce
the revisit time and increase the temporal resolution, as more ground is
covered in each pass.
Applications of Temporal Resolution
• Disaster Management:
High temporal resolution is essential for tracking rapidly changing
conditions during disasters like floods, wildfires, or hurricanes.
Example: MODIS for daily tracking of disaster zones.

55
 • Agriculture:
High temporal resolution helps monitor crop health, detect pests, and
track growth stages.
Example: Sentinel-2 for frequent monitoring of crop conditions.
• Climate Studies:
Temporal resolution aids in understanding long-term climate changes,
but may not require very frequent data collection.
Example: Landsat for monitoring climate patterns over decades.
• Urban Planning:
High temporal resolution helps track urban growth and changes in
infrastructure.
Example: MODIS or Sentinel-2 for monitoring city development.
Advantages of High Temporal Resolution
• Provides near-real-time data for dynamic monitoring (e.g., disaster
response or crop management).
• Essential for observing rapidly changing phenomena, like weather
patterns or environmental disasters.
• Enables frequent updates and detailed tracking of changes over time.
Limitations of Temporal Resolution
• High Cost: More frequent data capture may require advanced satellite
technology and increase operational costs.
• Large Data Volume: Frequent data collection generates large volumes of
data, requiring extensive storage and processing capabilities.
• Data Overload: Too much data can be overwhelming and may require
significant resources to analyze and extract useful information.
Conclusion
Temporal resolution plays a critical role in applications where frequent
monitoring and real-time data are necessary. High temporal resolution is
crucial for fast-changing events, such as natural disasters and crop health, while
lower temporal resolution may be more appropriate for long-term studies like
climate change or urban planning. Proper selection of temporal resolution is
essential for achieving the objectives of specific monitoring tasks.

56
Radiometric Resolution
Definition
Radiometric resolution refers to the ability of a sensor to measure the intensity
of the reflected or emitted energy from a target in discrete levels. It indicates
how precisely a sensor can detect variations in energy intensity (brightness) for
each pixel in an image. Higher radiometric resolution means the sensor can
distinguish finer differences in intensity, allowing for more detailed analysis of
variations in light.
For example:
• A sensor with 8-bit radiometric resolution can record 256 levels of
brightness (from 0 to 255).
• A sensor with 16-bit radiometric resolution can record 65,536 levels,
providing much more detailed intensity information.
Importance of Radiometric Resolution
• Capturing Detail: Higher radiometric resolution allows for a more
accurate representation of subtle differences in brightness, which is
important for detecting small changes in objects or areas.
• Application-Specific:
o High radiometric resolution is essential for detecting variations in
material types, moisture content, and subtle environmental
changes.
o Low radiometric resolution may suffice for broader applications
like land classification, where detailed intensity variations are not
as critical.
Types of Radiometric Resolution
• High Radiometric Resolution
o Definition: Captures a large number of intensity levels, typically 12
bits or more (e.g., 16-bit, 32-bit).
o Examples:
▪ WorldView-3: 16-bit radiometric resolution.
▪ Landsat-8: 12-bit radiometric resolution.

57
 o Applications:
▪ Precision agriculture, where small changes in crop health or
soil moisture are monitored.
▪ Environmental studies, such as detecting subtle variations in
vegetation or water quality.
▪ Remote sensing of urban areas, identifying different
materials (e.g., concrete vs. asphalt).
• Medium Radiometric Resolution
o Definition: Captures a moderate number of intensity levels,
typically 8 to 10 bits.
o Examples:
▪ Sentinel-2: 10-bit radiometric resolution.
▪ MODIS: 8-bit radiometric resolution for some bands.
o Applications:
▪ Land-use mapping and classification.
▪ Vegetation monitoring at a regional scale.
▪ Monitoring environmental changes over large areas.
• Low Radiometric Resolution
o Definition: Captures a smaller number of intensity levels, typically
4 to 6 bits.
o Examples:
▪ Older satellites or sensors with limited radiometric
capabilities.
o Applications:
▪ Large-scale land cover mapping where detailed intensity
differences are not as important.
▪ General climate monitoring and broad environmental
assessments.
Factors Affecting Radiometric Resolution

58
 • Sensor Design:
The sensor’s ability to differentiate energy levels depends on its detector
technology. More advanced sensors with better photodetectors can
capture finer variations in intensity.
• Bit Depth:
The bit depth of a sensor directly influences its radiometric resolution. A
sensor with a higher bit depth can represent more intensity levels. For
instance, an 8-bit sensor can represent 256 levels, while a 16-bit sensor
can represent 65,536 levels.
• Calibration:
Proper calibration of the sensor is essential to ensure accurate intensity
measurements. Miscalibrated sensors may lose their ability to detect
subtle intensity differences.
Applications of Radiometric Resolution
• Agriculture:
High radiometric resolution is useful for detecting subtle changes in crop
health, moisture levels, and stress due to pests or drought.
Example: Landsat-8 or WorldView-3 for precision farming applications.
• Environmental Monitoring:
High radiometric resolution helps in studying water quality, forest health,
and small-scale environmental changes.
Example: Sentinel-2 for monitoring forest vegetation or coastal
ecosystems.
• Urban Planning and Mapping:
Radiometric resolution helps in identifying different materials in urban
areas, such as distinguishing between buildings, roads, and vegetation.
Example: WorldView-3 for detailed urban infrastructure mapping.
• Disaster Management:
Higher radiometric resolution allows for more precise mapping of
affected areas after disasters like floods or wildfires, helping to assess
damage more accurately.
Example: High-resolution sensors like WorldView-3 for disaster
monitoring.
Advantages of High Radiometric Resolution

59
 • Provides more detailed data, allowing for precise differentiation between
materials and subtle variations.
• Essential for applications that require detecting small changes in light
intensity, such as precision agriculture or environmental monitoring.
• Allows for better classification of features and more accurate analysis in
various fields.
Limitations of Radiometric Resolution
• High Data Volume: Higher radiometric resolution generates larger
datasets, requiring more storage and processing power.
• Cost: Sensors with higher radiometric resolution are often more
expensive to develop and deploy.
• Data Overload: Too many intensity levels can sometimes make data
analysis more complex, requiring advanced techniques and tools to
process the information effectively.
Conclusion
Radiometric resolution plays a key role in determining the sensitivity of a
sensor to light intensity variations. High radiometric resolution is crucial for
applications that require detailed intensity analysis, such as monitoring crop
health or detecting environmental changes. While higher resolution provides
more precise data, it also brings challenges like larger data volumes and
increased processing costs. Proper choice of radiometric resolution depends on
the specific needs of the study or project.

60
Types of Sensors
Remote sensing sensors are devices that collect data by detecting
electromagnetic radiation (EMR) reflected or emitted from objects. These
sensors are classified based on how they operate and the type of data they
collect. Below are the classifications:

1. Classification Based on Energy Source

a. Passive Sensors
• Definition: Passive sensors depend on external energy sources like
sunlight to detect radiation reflected or emitted from objects.
• Examples:
o Multispectral Cameras (e.g., Landsat)
o Thermal Infrared Sensors (e.g., ASTER)
61
 • Applications:
o Land cover mapping
o Vegetation studies
o Surface temperature analysis
b. Active Sensors
• Definition: Active sensors emit their own energy and measure the
reflected signal from the target.
• Examples:
o RADAR (Radio Detection and Ranging)
o LiDAR (Light Detection and Ranging)
• Applications:
o Terrain mapping
o 3D modeling
o Forest canopy studies

2. Classification Based on Wavelength

a. Optical Sensors
• Definition: Optical sensors operate in the visible, near-infrared, and
shortwave infrared bands.
• Examples:
o Multispectral Sensors (e.g., Sentinel-2)
o Hyperspectral Sensors (e.g., Hyperion)
• Applications:
o Monitoring vegetation
o Water quality
o Mineral detection
b. Thermal Sensors

62
 • Definition: These sensors detect radiation in the thermal infrared
spectrum.
• Examples:
o ASTER Thermal Bands
o MODIS Thermal Sensors
• Applications:
o Surface temperature analysis
o Wildfire detection
c. Microwave Sensors
• Definition: Microwave sensors operate in the microwave region and can
penetrate clouds and vegetation.
• Examples:
o SAR (Synthetic Aperture Radar)
o RISAT (Radar Imaging Satellite)
• Applications:
o Flood monitoring
o Soil moisture analysis
o Glacier studies

3. Classification Based on Data Type

a. Imaging Sensors
• Definition: Imaging sensors capture data in the form of images.
• Examples:
o Multispectral Cameras
o SAR
• Applications:
o Urban planning

63
 o Forest monitoring
o Crop health analysis
b. Non-Imaging Sensors
• Definition: These sensors provide data in numerical or graphical formats
instead of images.
• Examples:
o Radiometers
o Spectroradiometers
• Applications:
o Atmospheric studies
o Oceanographic research

4. Classification Based on Platform

a. Spaceborne Sensors
• Definition: Mounted on satellites to observe Earth's surface.
• Examples:
o Landsat
o Sentinel
o MODIS
• Applications:
o Global monitoring of land, oceans, and atmosphere
b. Airborne Sensors
• Definition: Mounted on aircraft or drones for more localized studies.
• Examples:
o Hyperspectral Cameras
o LiDAR systems
• Applications:
64
 o High-resolution mapping
o Local studies
c. Ground-Based Sensors
• Definition: Stationary sensors used for localized monitoring on the
Earth's surface.
• Examples:
o Weather Radars
o Ground Spectroradiometers
• Applications:
o Atmospheric studies
o Field-based validation

Remote sensing platforms & sensors:

Remote sensing platform: ref page no. 8-9
Sensors: check previous one pg no. 61-65

IRS satellites:
Indian Remote Sensing Satellite (IRS)
Introduction
The IRS mission, initiated by the Government of India, aims to use satellite-
based remote sensing for natural resource management and sustainable
development. It comprises:
• Polar sun-synchronous satellites equipped with advanced sensors.
• Ground systems for data reception, processing, and dissemination.
• User-oriented data products for analysis and applications.
Objectives

65
 • Evaluate and manage natural resources like agriculture, hydrology,
geology, and forestry.
• Provide cost-effective and high-resolution data for various scientific and
practical applications.

Historical Development
1. IRS-1A (1988): India’s first operational remote sensing satellite.
2. IRS-1B (1991): Similar capabilities as IRS-1A.
3. IRS-1C (1995) & IRS-1D (1997):
o Introduced high-resolution sensors and improved revisit
capabilities.
4. IRS-P4 (1999):
o Specialized in oceanographic studies.

Technological Specifications
1. Sensor Systems
• LISS-I & LISS-II (IRS-1A/1B):
Purpose: Broad-area observation and land-cover mapping.
o Resolution: 72.5m (LISS-I), 36.25m (LISS-II).
o Spectral Bands: Visible and Near Infrared (NIR).
• LISS-III (IRS-1C/1D):
Purpose: Multispectral imaging for agriculture, forestry, and urban studies.
o Resolution: 21.2–23.5m (Visible & NIR), 63.6–70.5m (SWIR).
o Swath: 127–148km.
• Panchromatic Camera (PAN) (IRS-1C/1D):
Purpose: High-resolution imaging for urban planning, infrastructure
development, and land-use studies.

66
 o Resolution: 5.2–5.8m.
o Swath: 63–70km.
o Spectral Range: 0.50–0.75 microns.
o Steerable up to ±26° to improve revisit time (3–7 days).
• WiFS (Wide Field Sensor):
Purpose: Large-scale environmental studies like vegetation mapping,
snow cover, and soil moisture analysis.
o Resolution: 169–188m.
o Swath: 728–812km.
o Spectral Bands: 2 in visible and NIR.
• Ocean Colour Monitor (OCM)

Purpose: Oceanographic studies, including chlorophyll concentration,

coastal mapping, and fishery zone identification.
o Spatial Resolution: 360 meters.
o Swath Width: 1,420 kilometers.
o Spectral Bands: Eight narrow spectral bands in VNIR.

2. Orbit and Revisit Capability

• Altitude: ~780km (polar sun-synchronous orbit).
• Revisit Time:
o 3 days for PAN camera.
o 5–24 days for other sensors.
3. Payload and Onboard Features
• Onboard Tape Recorder: Stores limited sensor data.
• Programmable Payloads: Daily operation sequence loaded from ground
stations.
4. Ground Segment
• Telemetry, Tracking, and Command (TTC): Satellite tracking and control.

67
 • Data Processing & Dissemination: Converts raw satellite data into user-
friendly products.
Applications
1. Agriculture: Crop yield prediction, land use mapping.
2. Water Resources: Reservoir monitoring, watershed management.
3. Forestry: Forest cover assessment, deforestation monitoring.
4. Geology: Mapping mineral deposits, fault line identification.
5. Oceanography (IRS-P4): Monitoring coastal hydrodynamics, water
quality.
6. Urban Planning: Infrastructure development, land-use planning.

Significance
• Revolutionized natural resource management in India.
• Cost-effective solution for high-resolution and large-scale applications.
• Supports sustainable development and disaster management.
Special Features of IRS Satellites
• Onboard Tape Recorder (OBTR): Stores limited amounts of specified
sensor data.
• Programmable Operations: Daily payload operation sequences can be
loaded remotely.
• Telemetry Tracking and Command (TTC): Ground segment ensures
mission management, spacecraft operations, and scheduling

68
IRS satellites have transformed remote sensing in India by enabling cost-
effective, large-scale data collection for natural resource management,
environmental studies, and disaster response. Programs like LANDSAT and
SPOT have further strengthened their capabilities.

69
Landsat Satellite Programme
The Landsat Satellite Programme was initiated by NASA in collaboration with
the U.S. Department of the Interior. Its goal was to systematically acquire data
on Earth's resources via satellite. The first satellite, ERTS-1, was launched on
July 23, 1972, marking the beginning of what would become a long-term Earth
observation programme. The ERTS programme was later renamed Landsat to
differentiate it from other U.S. space programmes.
Mission Objective
The Landsat programme focuses on systematic, repetitive, medium-
resolution, multispectral data acquisition for Earth resource monitoring. These
satellites collect data that aid in the evaluation and management of Earth's
natural resources such as land use, agriculture, geology, and water bodies.
Key Sensors Used
The Landsat programme has utilized three main types of sensors across its
missions:
1. Return Beam Vidicon (RBV): Used in early missions (Landsat 1-3),
providing imagery for basic land-use studies.
2. Multispectral Scanner (MSS): Deployed in Landsat 1-5, capturing
multispectral images in 4 bands.
3. Thematic Mapper (TM): Used from Landsat 4 onwards, offering better
spatial and spectral resolution with 7 spectral bands, including thermal
infrared.
Orbit Details
• Orbit Altitude: Approximately 705 km (for most Landsat satellites).
• Revisit Time: Landsat satellites revisit the same area every 16 days.
• Orbit Type: Sun-synchronous, ensuring consistent lighting conditions for
every pass.
Applications
Landsat images have been crucial for a variety of fields, including:
• Agriculture: Crop health monitoring and land use management.

70
 • Forestry: Forest cover analysis, deforestation monitoring.
• Environmental Monitoring: Tracking natural disasters, soil erosion, and
pollution.
• Water Quality and Oceanography: Studying coastal areas, water bodies,
and ocean conditions.
• Urban Planning and Cartography: Land-use mapping, city expansion
tracking.
• Geology and Geography: Mineral exploration, terrain mapping.
Satellite Evolution
The Landsat programme has seen the launch of Landsat-1 to Landsat-5. Some
key missions include:
• Landsat-1 (1972): The first satellite to systematically monitor Earth
resources.
• Landsat-4 (1982): Introduced the Thematic Mapper with improved data
quality.
• Landsat-5 (1984-2013): Extended operational life, carrying TM for high-
resolution imaging.
The program later progressed to Landsat 7, Landsat 8, and Landsat 9, which
provide advanced imaging capabilities, including higher resolution and updated
spectral bands.
Data Resolution & Spatial Coverage
• RBV (Landsat 1-3): Spatial resolution of 80 m.
• MSS (Landsat 1-5): Spatial resolution of 79 m and a 185 km swath.
• TM (Landsat 4 and later): Spatial resolution of 30 m (6 bands), 120 m
(thermal infrared band).
The Landsat-8 and Landsat-9 satellites, launched more recently, offer an
enhanced resolution of 15 m for panchromatic bands, 30 m for multispectral
bands, and 100 m for thermal infrared.
Conclusion

71
The Landsat programme has become one of the most valuable resources for
Earth observation, providing decades of data used globally for environmental
monitoring, land management, and resource mapping. It continues to support
research and decision-making in agriculture, forestry, water management, and
urban development.

72
SPOT Satellite Programme
The SPOT (System Pour l'Observation de la Terre) Satellite Programme is a
collaborative effort involving France, Sweden, and Belgium, aimed at Earth
observation through high-resolution imaging. The first satellite, SPOT-1, was
launched on February 21, 1986, from the Kourou Launch Range in French
Guiana aboard an Ariane Launch Vehicle (AIV).
Key Features of SPOT Programme:
• First of its Kind: SPOT was the first Earth observation satellite system to
feature a linear array sensor using the push-broom scanning technique.
• Off-nadir Viewing: The push-broom technique enabled side-to-side off-
nadir viewing, which allowed for stereoscopic imaging—capturing the
same area from two different viewing points to create 3D images.
• Resolution: The satellite offers high-resolution imaging, making it
valuable for a variety of applications.
Sensors on SPOT Satellite:
1. Thematic Mapper (TM): Provides multispectral imaging for resource
management.
2. High Resolution Visible (HRV): Delivers high-resolution data for more
detailed observations.
Applications:
• Urban Planning: Helps assess urban growth, land use, and infrastructure.
• Transportation Planning: Aids in route planning, network management,
and urban transportation development.
• Natural Resources Management: Monitors land resources, agriculture,
forestry, and other environmental areas.
Satellite Specifications:
• SPOT satellites were designed for continuous Earth observation, enabling
revisit times as short as 1-3 days for high-resolution data collection.
• The SPOT satellites provided critical data for global environmental
monitoring and urban development, supporting a range of geospatial
applications across agriculture, forestry, and urban planning.

73
Conclusion:
The SPOT Programme revolutionized Earth observation with its off-nadir
viewing and stereoscopic imaging capabilities, significantly enhancing urban
planning and resource management. It has contributed valuable data for
decades, enabling global access to high-resolution imagery for a variety of
scientific, commercial, and governmental applications.

74
Additional information of SPOT
• The SPOT programme launched several satellites over time, such as
SPOT-1, SPOT-2, SPOT-3, SPOT-4, and SPOT-5, each with improved
resolution and technology.
• Each satellite had an operational life of about 5-7 years, providing
continuous monitoring of Earth’s surface.
• The SPOT project was a collaboration between France, Sweden, and
Belgium.
• SPOT satellite data was made available worldwide for scientific
research, mapping, and commercial use, helping with environmental
monitoring and urban planning.
• SPOT-5 provided 2.5 m resolution in panchromatic mode and 10 m
resolution in multispectral mode, allowing detailed images.
• The SPOT programme has been replaced by newer satellites, like
Pleiades, but the SPOT legacy still influences modern remote sensing.

NOTE : refer to spectrum for

concise answer regarding IRS topic
of our syllabus

PART I finished

75
 PART 2
TOPICS:
Remote Sensing Data Interpretation: Visual interpretation techniques, basic elements,
converging evidence, interpretation for terrain evaluation, spectral properties of soil, water
and vegetation.
Concepts of Digital image processing, image enhancements, qualitative & quantitative
analysis and pattern recognition, classification techniques and accuracy estimation .

Visual Interpretation Techniques

Definition:
• Visual interpretation involves manually analyzing and interpreting aerial
or satellite imagery using human perception to identify and classify
surface features.
• This technique relies on basic elements like tone, texture, shape, size,
pattern, shadow, and association.
Tools and Methods:
1. Stereoscopes: Used for 3D interpretation of overlapping aerial
photographs or satellite images.
2. False Color Composites (FCCs): Enhance specific features by assigning
different colors to various spectral bands.
3. Manual Digitization: Drawing polygons, lines, and points over imagery to
map features.
Let’s dive into deeper

1. Stereoscopes:
A stereoscope is a device used to view overlapping aerial photographs or
satellite images in 3D. When two photographs of the same area are taken
from slightly different angles, they overlap. By using a stereoscope, an
interpreter can view these overlapping images and perceive depth, which is
critical for analyzing topography and terrain features. The 3D view allows for
the identification of elevation changes, slopes, valleys, and ridges—
elements that would be difficult to discern in 2D images alone.

76
 • How it works: The stereoscope directs each eye to view one of the
overlapping images, creating the illusion of depth. This depth perception
is particularly useful for identifying features like buildings, mountains,
rivers, and forests.

The commonly used stereoscopes include:

i. Pocket Stereoscope
ii. Mirror Stereoscope

i. Pocket Stereoscope

• Most commonly used, made up of two convex lenses mounted on a

frame.
• Can be adjusted to accommodate various eye bases.
• How it works:
o The photographs are placed with corresponding images slightly
less than 2 inches apart, mimicking the natural eye spacing.

77
 2. Mirror Stereoscope
• Structure:
o Includes two large wing mirrors and two smaller eyepiece mirrors.
• How it works:
o Light rays from the points on the photograph (e.g., points a1, a2)
are reflected from the mirrors.
o The light rays are received by the eyes at a parallactic angle, which
the brain interprets to create a 3D effect.
o This process happens for every point in the image, resulting in a
3D view of the overlapping area.

Mirror Stereoscope

78
 Examples :
• Pocket Stereoscope: Hiking (used for viewing terrain in 3D during field
surveys)
• Mirror Stereoscope: Geology (used for analyzing rock formations in
3D from aerial images)

• Applications: Stereoscopic interpretation is widely used in topographic

mapping, landform analysis, and geological studies where the three-
dimensional layout is crucial.

2. False Color Composites (FCCs):

False Color Composites are a technique used in remote sensing to enhance
and highlight specific features in an image by assigning different colors to
various spectral bands beyond the visible spectrum (e.g., infrared,
ultraviolet). Since most satellite and aerial images capture a wider range of
electromagnetic spectrum bands, FCCs use these bands to provide more
detailed information than what is visible to the naked eye.
• How it works: In a typical false-color composite:
o Red, Green, and Blue (RGB) channels of an image are replaced
with non-visible spectrum bands, such as infrared (IR), near-
infrared (NIR), and shortwave infrared (SWIR).
o For instance, vegetation may appear in bright red in an FCC, which
helps highlight healthy vegetation against urban or barren land.
Similarly, water bodies might show up in shades of blue or black.
• Example: You’re looking at an image of a garden using an FCC. Healthy
plants appear red, while dry, wilting plants appear in shades of brown.
This lets you easily spot which plants need watering.
• Applications:
o Vegetation analysis: Healthy vegetation reflects high amounts of
near-infrared light, making it easily distinguishable from other land
covers in FCC.

79
 o Water body detection: FCC can make water bodies stand out more
clearly, especially in areas where water is not visible in natural
color.
o Urban planning: Urban areas, roads, and structures can be
differentiated from agricultural and natural landscapes.

False Colour Composite (FCC) of satellite data for Nagpur city, India

3. Manual Digitization:
Manual digitization involves the process of tracing and mapping features on a
digital image by drawing polygons, lines, or points. This technique is done by
hand, often using Geographic Information System (GIS) software or other
specialized mapping tools. The user "digitizes" features directly from the image
to create a vector-based representation of the features of interest.
• How it works:
o Polygons are used to outline irregular areas such as forest
boundaries or lakes.

80
 o Lines are used to map linear features such as roads, rivers, and
power lines.
o Points are used to mark specific locations, such as buildings or
landmarks.
o These digitized features are then used for creating detailed maps,
performing spatial analysis, or generating other geospatial data
products.
Example 1: You have a map of a small town with a satellite image. Using
manual digitization, you draw lines to outline the main roads, mark points for
schools, and create polygons around parks. This map can then be used for a
local traffic survey or to identify areas that need more green spaces.
Example 2: You have a satellite image of a lake. You use manual digitization to
trace the lake’s shoreline and mark the boundaries with a polygon. This can
later be used to study the lake's size or monitor changes in water levels
• Applications:
o Land use and land cover mapping: Manual digitization is a
traditional method for creating land use maps, which can be used
for urban planning, environmental monitoring, and resource
management.
o Disaster management: In cases of natural disasters like floods or
landslides, manual digitization can help create immediate maps
based on satellite or aerial imagery.
o Environmental monitoring: Mapping forest boundaries, wetlands,
or water bodies helps monitor changes over time.

81
 Manual digitization sample pic

Basic Elements
The process of interpreting aerial photographs and satellite images involves
analyzing certain characteristics of features on the image. These characteristics,
called fundamental picture elements, help in visual interpretation.
Below are key/basic visual interpretation techniques:
1. Tone
2. Texture
3. Shape
4. Size
5. Pattern
6. Location
7. Shadow
8. Association

1. Tone
• Definition: Refers to the lightness or darkness of the image or the color.
• Application: Helps in differentiating features such as water bodies (dark
tones), vegetation (green), urban areas (light gray or white), etc.
2. Texture
82
 • Definition: Refers to the visual pattern or arrangement of objects or
surfaces in the image.
• Application: Identifies surface roughness or smoothness, such as
distinguishing between water bodies (smooth) and forests (rough).
3. Shape
• Definition: Refers to the form or outline of objects in the image.
• Application: Used to identify specific features, such as roads (linear
shape), lakes (round or irregular), and buildings (rectangular).
4. Size
• Definition: Refers to the relative dimensions of objects in the image.
• Application: Helps in identifying features such as small buildings or fields
in agricultural landscapes.
5. Pattern
• Definition: Refers to the arrangement of objects in a regular or irregular
pattern.
• The well-known drainage patterns are: (i) Dendritic, (ii) Trellis, (iii)
Annualsr, (iv) Radial, (v) Rectangular, (vi) Parallel Type, (vii) Braided, (viii)
Anastomotic, (ix) Asymmetrical, (x) Collinear
• Drainage patterns take an important place amongst the various
interpretation elements, used as criteria for identification on geological
and geomorphological phenomena. Factors to be conSidered are
amongst other, the density which is measured for the erodibility of the
rocks; the amount of geological control on the drainage pattern and the
integration and homogeneity of the pattern.
• Application: Used to differentiate between natural features (like forests)
and man-made features (like urban areas).

83
 Main drainage
patterns

6. Location
• Definition: Refers to the position of objects in the image.
• Application: Helps in identifying specific regions, such as rivers,
mountain ranges, or cities by their spatial relationship in the image.
7. Shadow
• Definition: Refers to the dark areas caused by obstacles casting shadows.
• Application: Useful in interpreting topography, height of objects (like
buildings), and estimating the slope of terrain.
8. Association
• Definition: Refers to the relationship between different objects and
features within the image.
• Application: Identifies patterns or combinations of features, such as
vegetation near water bodies, or cities near highways.

84
These techniques are often used together to provide a comprehensive
understanding of the image and its features, facilitating applications like land
use mapping, vegetation studies, urban planning, and environmental
monitoring.

Application of Aerial Photo Interpretation

Aerial photos have been used for several applications. Using the principles of
visual photo interpretation, information about the earth's surface features, can
be obtained. Some of the areas where aerial photos have been extensively
used are:
(i) Topographical mapping,
(ii) Geology,
(iii) Soil mapping,
(iv) Forestry,
(v) Terrain evaluation,
(vi) Land use/Land cover mapping,
(vii) Agriculture,
(viii) Water resources, and
(ix) Environmental studies/flood damage studies.
1. Topographical Mapping:
o Aerial photos are used to create topographical maps by using the
overlap between successive photos.
o A stereo projector recreates the three-dimensional model of the
ground using the overlap.
o Aerial mapping is more cost-effective and efficient compared to
traditional ground survey methods, requiring fewer control points.
2. Geology:
o Aerial photos provide a synoptic, bird’s-eye view of terrain, useful
for identifying geological features like faults, rock types, erosion
resistance, and drainage patterns.
o Lineament effects in aerial photos help identify geological features
that may be difficult to detect from the ground.

85
 o ' it is possible to do geological mapping for the following:
(i) The photographic tone of the rock body to that of
adjacent rocks
(ii) resistance to erosion,
(iii) boundaries of rock types,
(iv) topographic expression,
(v) boundaries of outcrops,
(vi) the joints and their pattern,
(vii) faults and their pattern,
(viii) drainage pattern
(ix) the vegetation cover, and
(x) the lineament pattern.
3. Soil Mapping:
o Aerial photos help in soil mapping by correlating the type of soil
with the tone and texture observed in the photos.
o Physiographic systems, which include conditions like stratigraphy
and geomorphology, are inferred using aerial photos for soil
mapping.
o Models based on physiographic systems are used to predict soil
patterns and are validated through field checks.
4. Forestry:
o Aerial photos are used to identify tree species, estimate tree
density, and measure tree height and volume.
o Tree species have characteristic shapes, sizes, and patterns visible
in photos, which can be linked to physiological and environmental
factors like moisture levels.
5. Terrain Evaluation:
o Aerial photos are used to gather information about the ground for
various uses, such as military planning, mineral resource
development, and groundwater exploration.
o Terrain classification is based on surface configurations, surficial
deposits, land use, and vegetation.

86
6. Land Use/Land Cover:
o Aerial photos can delineate land use (human activities) and land
cover (natural features) with high precision.
o Different classification levels help identify various land uses like
residential, commercial, industrial, and agricultural areas.

7. Agricultural Applications:
o Aerial photos help assess crop conditions, classify crop types, and
estimate crop yields.
o The photos can reveal crop health and possible damage due to
pests, water issues, and nutrient deficiencies.
8. Water Resources:
o Aerial photos are used for mapping surface water bodies like lakes
and rivers and assessing groundwater potential.
o Surface water bodies appear darker due to the absorption of light
by water.
o Features like springs, wells, and vegetation patterns help identify
groundwater zones.
9. Environmental and Flood Damage Studies:

87
 o Aerial photos are used to study environmental issues such as
water pollution, deforestation, and industrial pollution.
o Water pollution can be identified through different tones in aerial
photos, with pure water appearing light and polluted water
appearing darker.
o Flood damage can be assessed by comparing aerial photos taken
before and after floods, highlighting affected areas.

Interpretation for Terrain evaluation

Terrain Evaluation: Terrain evaluation is a crucial study for effective land
planning and utilization. It aims to gather and systematically evaluate
information about the land for various purposes such as:
• Military requirements: Short-term evaluations for strategic military needs
in specific zones.
• Peace-time needs: Long-term projects like mineral exploration,
construction material availability, groundwater exploration, and land
development.
The ultimate goal is to assess the land in a way that meets the specific needs of
different users, whether for defense, agriculture, construction, or natural
resource management. For inaccessible areas, aerial photos become the
primary tool, providing a valuable means of exploring and understanding terrain
without physical access. The aerial photos, especially when viewed through a
stereoscope, allow the "ground" to be examined as if it were brought into the
laboratory, offering a detailed view of the landscape.
Classification Units in Terrain Evaluation: Two key units used to classify and
evaluate terrain are Pattern and Facet.
1. Facet:
o It is the fundamental unit in the classification system.
o A facet refers to a section of the land that has uniform physical
characteristics. It is classified based on:
▪ Surface configuration (shape of the land)

88
 ▪ Nature of surficial deposits (soil and material on the surface)
▪ Water characteristics (surface and subsurface water regions)
▪ Land use and how it interacts with other terrain units
2. Pattern:
o This refers to areas that have a recurring arrangement of
topography and surface materials, including soils.
o It is recognized based on geological setup, climate, and topography.
Information Recorded for Terrain Units: For each terrain unit, the following
information is recorded:
• Pattern Card: Describes the pattern, climate, and other relevant data.
• Facet Card: Includes information about the ground, surficial deposits, soil
properties, land use, vegetation, water supply, engineering resources, and
more.
Terrain Study Methodology:
1. Library Work:
o Collect basic material and background information about the area
to understand the environment before analysis.
2. Laboratory Exercise:
o Study air photos for quality, scale, and map features.
o Identify and delineate major landscape patterns and facets.
3. Ground Reconnaissance:
o Conduct preliminary field visits to gather more data on-site.
4. Study of Aerial Photos:
o Delineate facets and plan detailed fieldwork based on aerial photo
interpretation.
5. Detailed Field Work:
o Check photo characteristics, record terrain data, collect soil
samples, conduct field tests for moisture and soil strength.

89
 o Finalize the list of patterns and facets.
6. Laboratory Tests:
o Perform soil tests, rock identification, and other analyses.
7. Completion of Reports:
o Finalize the report, including appendices, information cards, and
facet maps.
o A confirmatory reconnaissance (field check) may be conducted to
verify findings.
Photo Interpretation Process: Photo interpretation in terrain studies follows a
structured approach:
• Observation and Measurements: Look at photos closely to identify
features.
• Logical Reasoning: Deduce possible explanations for the observed
patterns.
• Influence: Consider the potential impact of these patterns on terrain
characteristics.
Equipment Required:
• Pocket Stereoscope: For quick scanning of aerial photos in the field.
• Mirror Stereoscope: For detailed work and a more accurate view of
photos.
• Parallax Bar: Used to measure height and elevation in terrain.
• Sketch Master: Transfers the interpreted data onto base maps or
existing toposheets.
Recognition Elements for Photo Interpretation: When interpreting aerial photos,
several key elements help describe and differentiate terrain units:
1. Position in Landscape: How the terrain unit fits into the broader
landscape and its relationship with surrounding areas.
2. Morphology and Surface Configuration: The shape and structure of the
land, including small-scale features.

90
 3. Drainage Patterns: Observing water flow, texture, and density of drainage
systems in the area.
4. Erosional Features: Identifying areas with erosion, such as gullies or worn-
out sections of land.
5. Land Use and Cultural Features: The presence of human activity, like
agriculture, settlements, or infrastructure.
6. Vegetation: Types of plant cover that can be seen from aerial photos,
indicating soil health and climate.
7. Tonal Variation and Photo Textures: Changes in photo tone and textures
that might indicate different terrain features or soil types.
Through these methods and tools, terrain evaluation and aerial photo
interpretation provide a comprehensive understanding of the land's
characteristics and suitability for various uses, even in remote and inaccessible
areas.

NOTE : above info is from GPT

below is from Txt bk
I’m mentioning both ,so choose yourself although the
same content but organization is different that’s it

91
92
93
CONVERGENCE OF EVIDENCE

94
Spectral Properties of Soil, Water, and Vegetation
In remote sensing, the spectral properties of soil, water, and
vegetation are crucial for understanding their characteristics and for
differentiating them in satellite and aerial images. These properties
refer to how these elements interact with electromagnetic radiation,
particularly across different wavelengths.
1. Spectral Properties of Soil:
• Reflectance: Soils have unique spectral reflectance patterns
based on their composition, texture, moisture content, and
organic matter. Generally, soils reflect more in the infrared (IR)
and less in the visible spectrum.
• Visible Spectrum (400-700 nm): In the visible range, soils reflect
less sunlight compared to vegetation. Dark soils tend to absorb
more light, while lighter soils reflect more.
• Near-Infrared (NIR, 700-1100 nm): Soil reflects moderately in
the NIR range, with the reflectance varying based on soil type
and moisture content. Wet soils have lower reflectance in NIR,
while dry soils have higher reflectance.
• Shortwave Infrared (SWIR, 1100-2500 nm): The reflectance of
soil increases in this range, and the difference in reflectance can
help distinguish soil moisture content and composition. Soils
with higher clay or organic matter content tend to have higher
reflectance in SWIR.
• Key Factors Influencing Soil Spectral Properties:
o Moisture content
o Organic material and vegetation cover
o Soil texture (e.g., clay, sand, silt)
2. Spectral Properties of Water:

95
 • Visible Spectrum (400-700 nm): Water absorbs most of the light
in the visible spectrum, which is why it appears blue or green
depending on the depth and impurities in the water. The
reflectance is generally low in the visible range, except in shallow
waters where it may reflect light from the bottom.
• Near-Infrared (NIR, 700-1100 nm): Water absorbs almost all NIR
radiation. This strong absorption is a key property for
distinguishing water from other land features in remote sensing
images. Water appears dark or black in the NIR range.
• Shortwave Infrared (SWIR, 1100-2500 nm): Water continues to
absorb strongly in the SWIR range, making it even easier to
distinguish from land surfaces in this band.
• Key Factors Influencing Water Spectral Properties:
o Depth of water
o Presence of suspended particles or dissolved substances
o Turbidity and color (e.g., clear water vs. polluted water)
3. Spectral Properties of Vegetation:
• Visible Spectrum (400-700 nm): Vegetation primarily absorbs
red (around 660 nm) and blue light (around 450 nm) for
photosynthesis, and reflects green light (around 550 nm), which
is why vegetation appears green to the human eye. The
chlorophyll pigments are responsible for this absorption and
reflection.
• Near-Infrared (NIR, 700-1100 nm): Healthy vegetation reflects a
high amount of NIR radiation due to the internal structure of the
leaves, which scatter light. This high NIR reflectance is a critical
feature for distinguishing vegetation from other land cover types
in remote sensing images.

96
• Shortwave Infrared (SWIR, 1100-2500 nm): In this range,
vegetation reflectance decreases as the leaf water content
increases, with higher moisture content resulting in lower
reflectance. This range helps assess vegetation health and
moisture content.
• Key Factors Influencing Vegetation Spectral Properties:
o Leaf pigments (chlorophyll)
o Leaf structure and health (e.g., senescence or disease)
o Water content in the vegetation

97
Digital Image Processing
Digital Image Processing (DIP)
Digital Image Processing refers to the use of computer algorithms and
techniques to manipulate and enhance images in digital form. This process
involves analyzing, modifying, and extracting information from digital images
using mathematical and computational methods. The goal of DIP is to improve
the quality of the image or to extract useful information from it, and it finds
applications in various fields like remote sensing, medical imaging, industrial
inspection, and computer vision.
Major Steps in Digital Image Processing
[Link] Acquisition:
• Collecting raw data from remote sensing platforms (satellites, drones,
etc.).
• Data is often in the form of digital numbers (DN) representing brightness
values.
2 Preprocessing:

98
• Radiometric Correction: Corrects sensor or atmospheric distortions (e.g.,
haze, noise).
• Geometric Correction: Aligns the image with real-world coordinates.
3 Image Enhancement:
• Improves the visual quality of images to make features more
distinguishable.
• Techniques include:
• Contrast Stretching: Enhances contrast by spreading DN values.
• Spatial Filtering: Highlights edges or smoothens data.
• False Color Composite (FCC): Assigns colors to specific spectral bands for
better feature differentiation.
4 Image Classification:
• Assigns pixels to specific categories based on their spectral properties.
• Types:
• Supervised Classification: Uses training data to classify pixels.
• Unsupervised Classification: Groups pixels into clusters based on
similarity without prior knowledge.
5 Feature Extraction:
• Identifies specific patterns or features (e.g., roads, water bodies) using
advanced algorithms.
6 Post-Processing:
• Final adjustments to improve accuracy (e.g., smoothing classified
outputs, validation).
7 Accuracy Assessment:
• Validates the results of classification using reference data.
Key Techniques in Digital Image Processing
1 Image Enhancement:
• Improves interpretability of data.

99
 • Examples: Histogram equalization, band ratioing.
2 Image Transformation:
• Converts images for specific applications.
• Examples: Principal Component Analysis (PCA), NDVI for vegetation
analysis.
3 Pattern Recognition:
• Automates feature detection using machine learning or deep learning
techniques.
4 Image Classification:
• Grouping image pixels into meaningful classes.
• Examples: Land use/land cover maps.
Applications of Digital Image Processing
• Environmental Monitoring: Analyzing land degradation, pollution, and
deforestation.
• Urban Planning: Mapping urban growth and infrastructure.
• Agriculture: Monitoring crop health, yield, and soil conditions.
• Disaster Management: Mapping flood zones, fires, and earthquake-
affected areas.
• Water Resources: Monitoring water quality and detecting reservoirs.

Image Enhancements
Definition
• Digital image enhancement refers to mathematical operations applied to
remote sensing data to improve the visual appearance of an image for
better interpretation. It amplifies differences between features, increasing
contrast and making the image more interpretable.
Categories of Image Enhancement
1 Point Operations:

100
 • Modify the values of each pixel independently.
• Includes techniques like contrast enhancement and band
combinations.
2 Local Operations:
• Modify the value of a pixel based on its surrounding pixel values.
• Includes techniques like spatial filtering.
Contrast Enhancement Techniques
1 Linear Contrast Stretching:
• Expands the narrow range of brightness values (digital numbers) to
utilize the entire dynamic range of the sensor.
• Formula: Y=a+bXY = a + bXY=a+bX (where XXX is the input gray value,
YYY is the output gray value, and aaa and bbb are constants).
2 Histogram Equalization:
• Adjusts the image histogram to create uniform distribution, increasing
contrast across all brightness levels.
3 Logarithmic and Exponential Enhancements:
• Logarithmic: Highlights darker regions by compressing high brightness
values.
• Exponential: Highlights brighter areas while compressing low brightness
values.
4 Gaussian Stretching:
• Converts input pixel values to a normal distribution, enhancing details
based on Gaussian laws.
Spatial Filtering Techniques
1 Low-Pass Filtering:
• Smoothens the image by reducing noise and enhancing uniform
regions.
• Example: Mean and median filters.
2 High-Pass Filtering:

101
 • Enhances edges and fine details by removing low-frequency
components.
• Example: Laplacian and Roberts filters.
3 Edge Enhancement:
Sharpens boundaries between features using high-frequency filters
like Laplacian edge filtering.
Applications of Image Enhancements
• Agriculture: Crop health monitoring.
• Urban Planning: Infrastructure mapping.
• Geology: Fault and lineament detection.
• Disaster Management: Highlighting flood and wildfire zones

Qualitative & quantitative analysis and pattern

recognition
• Qualitative Analysis: Focuses on the interpretation of visible
patterns, textures, and features based on human perception and
contextual knowledge.
• Quantitative Analysis: Focuses on extracting measurable,
numerical data from images, using statistical, radiometric, or
mathematical techniques to provide an objective understanding
of the image.
Qualitative analysis involves the interpretation of remote sensing
data based on visual inspection and understanding of patterns,
textures, and general features in an image. In the context of DIP,
qualitative analysis focuses on understanding the image's visual
characteristics and how the spatial features are represented without
relying on numerical measures.
Key Aspects:
1. Visual Interpretation:

102
 o Remote sensing images are often interpreted by experts
who look at the color, texture, patterns, and shape of the
features in the image. For example, identifying land cover
types (forest, water bodies, urban areas) based on visual
cues.
2. Textural Features:
o Textural analysis focuses on identifying patterns or
textures in an image, like regularity or randomness in the
spatial arrangement of pixels. This can help in identifying
specific objects or land features (e.g., vegetation, roads).
3. Feature Recognition:
o Features such as roads, rivers, and urban areas are
identified based on their visual characteristics, such as
linearity, continuity, and color contrast. These features are
recognized without the need for extensive numerical
analysis.
4. Contextual Analysis:
o This type of analysis involves understanding the context of
image features, such as topography, climate, or land use,
to better interpret what the features represent (e.g.,
recognizing a hill based on its shading and surrounding
vegetation).
Applications in DIP:
• Pattern Recognition: Identifying land cover types or objects
using visual cues.
• Image Classification: Grouping pixels into categories based on
qualitative differences in their appearance.

103
 • Land Use Mapping: Creating maps that reflect land use
categories, based on observed patterns and textures in the
image.
Quantitative analysis in DIP involves the use of mathematical and
statistical methods to analyze the pixel values of an image and
derive measurable data. In RS/GIS, quantitative analysis is used to
objectively assess features and derive specific, measurable
information, often in the form of numerical values.
Key Aspects:
1. Statistical Analysis:
o Statistical measures such as mean, standard deviation,
and variance are applied to the pixel values to understand
the image's distribution. For example, analyzing the
brightness levels in satellite imagery to classify different
land cover types.
2. Radiometric Analysis:
o This involves the analysis of pixel values based on their
intensity (brightness), which are then used to assess
physical properties like temperature, vegetation health,
and soil moisture. Quantitative methods help in
distinguishing different materials based on their
radiometric signatures.
3. Classification Techniques:
o Quantitative analysis is often used in supervised and
unsupervised classification algorithms, where statistical
methods like k-means clustering, maximum likelihood, or
decision trees are applied to categorize pixels based on
spectral signatures.
4. Change Detection:

104
 o By comparing multiple images (e.g., before and after a
natural disaster), quantitative methods like image
differencing or post-classification comparison help identify
areas of change (e.g., urban expansion, deforestation).
5. Geospatial Analysis:
o Quantitative analysis in GIS involves spatial data analysis
like overlay analysis, buffering, and distance calculation.
This helps in deriving accurate measurements of distance,
area, or spatial relationships between features.
Applications in DIP:
• Vegetation Indices: Calculating vegetation indices like NDVI
(Normalized Difference Vegetation Index) to assess plant health.
• Image Classification: Using algorithms like k-means or neural
networks to classify satellite imagery into specific land cover
categories.
• Change Detection: Quantifying the changes in a region over
time, such as urban growth or deforestation.
• Surface Temperature Mapping: Using infrared data to calculate
temperature variations in an area.

Pattern Recognition
Definition:
Pattern recognition in the context of remote sensing and GIS
involves the automated or manual identification and classification of
spatial patterns in images. It uses Digital Image Processing (DIP)
techniques to recognize and categorize features or objects.
Types of Pattern Recognition:
1. Supervised Classification:
o Uses pre-defined training data to classify image pixels.
105
 o Example: Classifying satellite imagery into land cover
types (e.g., forest, water, urban).
o Requires labeled data for training.
2. Unsupervised Classification:
o Clusters pixels into groups based on similarity without
prior knowledge.
o Example: Identifying unknown land cover categories or
anomalies.
o Algorithms like K-means and ISODATA are commonly used.
3. Object-Based Classification:
o Focuses on grouping neighboring pixels into meaningful
objects.
o Example: Detecting buildings or roads in high-resolution
imagery.
4. Feature-Based Recognition:
o Identifies specific features such as lines, edges, and
shapes.
o Example: Recognizing drainage networks or fault lines in
geological mapping.
Process of Pattern Recognition:
1. Preprocessing: Apply image enhancements to improve feature
visibility.
2. Feature Extraction: Extract characteristics like shape, texture, or
spectral values.
3. Classification: Use algorithms to group or label the features.
4. Validation: Verify the results against ground truth or reference
data.
106
 Applications:
• Land cover and land use mapping.
• Urban growth and sprawl analysis.
• Detecting changes in vegetation or water bodies.
• Automated disaster monitoring, such as flood extent mapping.
By combining qualitative and quantitative analyses with pattern
recognition, RS & GIS enables detailed and accurate interpretation
of spatial data for diverse applications.

Image Classification in Remote Sensing and GIS

Image classification is the process of automatically categorizing pixels
in an image into distinct land cover classes based on the spectral or
spatial patterns present. This classification process is fundamental to
understanding and mapping different features of the Earth's surface.
There are two main types of image classification:
• Spectral Pattern Recognition: This method uses the spectral
information of each pixel (from multispectral data) to classify
the pixels into predefined categories.
• Spatial Pattern Recognition: This method categorizes pixels
based on the spatial relationships and context of the
surrounding pixels.
Image classification falls under the broader category of Pattern
Recognition, where the objective is to identify patterns associated
with each pixel in the image in terms of the characteristics of the
surface features.
Supervised Classification

107
In supervised classification, the analyst provides "training data,"
which are pixel samples already known to belong to specific classes
(e.g., water, forest, urban). The algorithm then classifies all other
pixels based on how closely they match the spectral characteristics of
the training data.
Key Stages in Supervised Classification:
1. Training Stage: The analyst selects training areas that represent
various land cover types in the image and develops numerical
descriptions of the spectral characteristics of each class.
2. Classification Stage: Each pixel is assigned to the class it most
closely resembles. If a pixel does not closely match any training
data, it may be labeled as "Unknown."
3. Output Stage: The results are presented in the form of thematic
maps, tables, or digital data files, which can be further used in
GIS for spatial analysis.
Supervised classification methods include algorithms such as:
• Minimum Distance to Means
• Maximum Likelihood Method
• Decision Tree Classification
• Bayesian Method
• Parallelepiped Method
Training Dataset:
A training dataset is a collection of pixels whose class membership is
already known. The quality and selection of training points
significantly impact the accuracy of the classification. A few
guidelines for selecting training points include:

108
 • Sufficient Quantity: Select enough points for each class to
account for variability (at least 10 * N, where N is the number
of features).
• Representative Sampling: Choose training points that represent
the typical and varied characteristics of each class across the
entire image.
• Unimodal Distribution: Ensure the training points have a
unimodal histogram, as bimodal histograms may indicate
multiple classes within the same training region.
• Evaluation: Half of the training data should be used for training,
and the other half should be used for testing the accuracy of
the classifier.

Unsupervised Classification
In unsupervised classification, the algorithm does not rely on pre-
labeled training data. Instead, it divides the image into classes based
on inherent data properties. It assumes that pixels in the same class
are similar, while pixels in different classes are distinct.
Process of Unsupervised Classification:
1. Data Clustering: The algorithm groups data into clusters based
on similarities.
2. Spectral Classes: The resulting clusters are called "spectral
classes," which might not directly correspond to real-world land
cover classes (e.g., "forest" or "water").
3. Class Merging and Renaming: The analyst typically merges or
relabels these spectral classes into meaningful information
classes based on their characteristics and known land cover
types.

109
Unsupervised classification is useful when there is little or no prior
knowledge of the image content but may require post-processing to
interpret the results meaningfully.

Separability of Classes
The separability of classes refers to how distinct different classes are
in the image's feature space. High separability means that classes
have well-defined boundaries and are easily distinguishable. Low
separability means classes may overlap and are harder to
differentiate. In practice, the more distinct the classes are, the more
accurate the classification will be.
Conclusion
In summary, image classification is a vital tool in remote sensing for
analyzing and categorizing terrain into land cover classes. Whether
using supervised or unsupervised methods, the process involves
understanding spectral and spatial patterns and applying appropriate
algorithms to produce reliable outputs for further analysis in GIS
applications.

110
111
Accuracy Estimation

112
113
Conclusion:
Accuracy in image processing is crucial to assess the performance of
classification models. Several metrics and methods like confusion
matrices, Kappa coefficient, and cross-validation help estimate and
fine-tune the performance of image classifiers. Ensuring high
accuracy often involves addressing data quality, algorithm choices,
and feature selection, alongside continuous evaluation and testing.

114
115
116
117
END

118