BASIC CONCEPTS OF REMOTE SENSING

1. Introduction

Remote sensing is an art and science of obtaining information about an object or

feature without physically coming in contact with that object or feature. Humans
apply remote sensing in their day-to-day business, through vision, hearing and sense of
smell. The data collected can be of many forms: variations in acoustic wave
distributions (e.g., sonar), variations in force distributions (e.g., gravity meter),
variations in electromagnetic energy distributions (e.g., eye) etc. These remotely
collected data through various sensors may be analyzed to obtain information about the
objects or features under investigation. In this course we will deal with remote sensing
through electromagnetic energy sensors only.
What are the main effects of atmosphere on the EMR between the
sources and objects? Use sketches to support your answer.

Thus, remote sensing is the process of inferring surface parameters from measurements
of the electromagnetic radiation (EMR) from the Earth’s surface. This EMR can either
be reflected or emitted from the Earth’s surface. In other words, remote sensing is
detecting and measuring electromagnetic (EM) energy emanating or reflected from
distant objects made of various materials, so that we can identify and categorize
these objects by class or type, substance and spatial distribution.

Fig. 1. Schematic representation of remote sensing technique

Remote sensing provides a means of observing large areas at finer spatial and
temporal frequencies. It finds extensive applications in civil engineering including
watershed studies, hydrological states and fluxes simulation, hydrological modeling,
disaster management services such as flood and drought warning and monitoring, damage
assessment in case of natural calamities, environmental monitoring, urban planning etc.

Basic concepts of remote sensing are introduced below.

2. Electromagnetic Energy

Electromagnetic energy or electromagnetic radiation (EMR) is the energy propagated in the

form of an advancing interaction between electric and magnetic fields. It travels with the
velocity of light. Visible light, ultraviolet rays, infrared rays, heat, radio waves, X-rays all
are different forms of electro-magnetic energy.

Electro-magnetic energy (E) can be expressed either in terms of frequency (f) or wave length
(λ) of radiation as

E=hcf or hc/λ (1)

where h is Planck's constant (6.626 x 10-34 Joules-sec), c is a constant that expresses the
celerity or speed of light (3 x 108 m/sec), f is frequency expressed in Hertz and λ is the
wavelength expressed in micro meters (1µm = 10-6 m).

As can be observed from equation (1), shorter wavelengths have higher energy content and
longer wavelengths have lower energy content.

Distribution of the continuum of energy can be plotted as a function of wavelength (or

frequency) and is known as the EMR spectrum (Fig. 2).
 Thermal infrared
Ultraviolet rays

Near Infrared
Gamma rays

Radio waves
Visible light

Microwave
Wave 10-6 10-5 X rays
10-4 10-3 10-2 10-1 1 10 102 103 104 105 106 107 108 109
length (μm)

Violet Blue Green Yellow Orange Red

Near
Ultraviolet
Infrared
Wavelength 0.40 0.48 0.54 0.58 0.60 0.65
(μm)

Fig. 2. Electromagnetic radiation spectrum

In remote sensing terminology, electromagnetic energy is generally expressed in terms of

wavelength, λ.

All matters reflect, emit or radiate a range of electromagnetic energy, depending upon
the material characteristics. In remote sensing, it is the measurement of electromagnetic
radiation reflected or emitted from an object, is the used to identify the target and to
infer its properties.

3. Principles of Remote Sensing

Different objects reflect or emit different amounts of energy in different bands

of the electromagnetic spectrum. The amount of energy reflected or emitted
depends on the properties of both the material and the incident energy (angle of
incidence, intensity and wavelength). Detection and discrimination of objects or surface
features is done through the uniqueness of the reflected or emitted electromagnetic
radiation from the object.

A device to detect this reflected or emitted electro-magnetic radiation from an object is

called a “sensor” (e.g., cameras and scanners).
A vehicle used to carry the sensor is called a “platform” (e.g., aircrafts and
satellites).
 What are the main stages in
remote sensing? List or sketch.
Main stages in remote sensing are the following.

A. Emission of electromagnetic radiation

 The Sun or an EMR source located on the platform
B. Transmission of energy from the source to the object
 Absorption and scattering of the EMR while transmission
C. Interaction of EMR with the object and subsequent reflection and emission
D. Transmission of energy from the object to the sensor
E. Recording of energy by the sensor
 Photographic or non-photographic sensors
F. Transmission of the recorded information to the ground station
G. Processing of the data into digital or hard copy image
H. Analysis of data

These stages are shown in Fig. 3.

Fig.3 Important stages in remote sensing

What is the main difference between Passive and Active Remote
Sensing? Describe or sketch.

4. Passive/ Active Remote Sensing

Depending on the source of electromagnetic energy, remote sensing can be classified

as passive or active remote sensing.

In the case of passive remote sensing, source of energy is that naturally available such as
the Sun. Most of the remote sensing systems work in passive mode using solar energy as
the source of EMR. Solar energy reflected by the targets at specific wavelength bands
are recorded using sensors on board air-borne or space borne platforms. In order to ensure
ample signal strength received at the sensor, wavelength / energy bands capable of
traversing through the atmosphere, without significant loss through atmospheric
interactions, are generally used in remote sensing

Any object which is at a temperature above 0o K (Kelvin) emits some radiation, which is
approximately proportional to the fourth power of the temperature of the object. Thus the
Earth also emits some radiation since its ambient temperature is about 300o K. Passive
sensors can also be used to measure the Earth’s radiance but they are not very popular
as the energy content is very low.

In the case of active remote sensing, energy is generated and sent from the remote
sensing platform towards the targets. The energy reflected back from the targets are
recorded using sensors on board the remote sensing platform. Most of the microwave
remote sensing is done through active remote sensing.

As a simple analogy, passive remote sensing is similar to taking a picture with an ordinary
camera whereas active remote sensing is analogous to taking a picture with camera having
built-in flash (Fig. 5).
 Fig. 5 Schematic representation of passive and active remote sensing

5. Remote Sensing Platforms List three classifications of platform in remote sensing.

Remote sensing platforms can be classified as follows, based on the elevation from the
Earth’s surface at which these platforms are placed.

 Ground level remote sensing

o Ground level remote sensors are very close to the ground
o They are basically used to develop and calibrate sensors for different features
on the Earth’s surface.
 Aerial remote sensing
o Low altitude aerial remote sensing
o High altitude aerial remote sensing
 Space borne remote sensing
o Space shuttles
o Polar orbiting satellites
o Geo-stationary satellites

From each of these platforms, remote sensing can be done either in passive or active mode.
 Fig. 6. Remote sensing platforms

6. Airborne and Space-borne Remote Sensing

In airborne remote sensing, downward or sideward looking sensors mounted on aircrafts are
used to obtain images of the earth's surface. Very high spatial resolution images (20 cm or
less) can be obtained through this. However, it is not suitable to map a large area.
Less coverage area and high cost per unit area of ground coverage are the major
disadvantages of airborne remote sensing. While airborne remote sensing missions
are mainly one-time operations, space-borne missions offer continuous monitoring of
the earth features.

LiDAR, analog aerial photography, videography, thermal imagery and digital photography
are commonly used in airborne remote sensing.

In space-borne remote sensing, sensors mounted on space shuttles or satellites orbiting the
Earth are used. There are several remote sensing satellites (Geostationary and Polar orbiting)
providing imagery for research and operational applications.
9. Advantages and Disadvantages of Remote Sensing

Advantages of remote sensing are:

a) Provides data of large areas

b) Provides data of very remote and inaccessible regions
c) Able to obtain imagery of any area over a continuous period of time through which the
any anthropogenic or natural changes in the landscape can be analyzed
d) Relatively inexpensive when compared to employing a team of surveyors
e) Easy and rapid collection of data
f) Rapid production of maps for interpretation

Disadvantages of remote sensing are:

a) The interpretation of imagery requires a certain skill level

b) Needs cross verification with ground (field) survey data
c) Data from multiple sources may create confusion
d) Objects can be misclassified or confused
e) Distortions may occur in an image due to the relative motion of sensor and source
What are the causes for the reduction of EMR energy
between the
sources and objects?

EMR energy is reduced due to

absorption,
scattering,
reflection
, and distance.
• Remote sensing is the science and art of acquiring and analyzing information about
objects or phenomena from a distance.

• As humans, we are familiar with remote sensing in that we rely on visual

perception to provide us with much of the information about our surroundings. As
sensors, however, our eyes are greatly limited by

1) sensitivity to only the visible range of electromagnetic energy;

2) viewing perspectives dictated by the location of our bodies; and
3) the inability to form a lasting record of what we view.

Because of these limitations, humans have continuously sought to develop

technology that would enable us to increase our ability to see and record the
physical properties of our environment.

WHY REMOTE SENSING?

• Satellite remote sensing, with its view
of the earth’s features, regular repetitive
coverage over large areas, and digital mode of
data capture, offers :
1. An effective means of inventorying and
updating natural resources information,
2. Monitoring land use changes and
environmental impacts near real time, and
3. Providing a historical profile for the purpose
of policy formulation. ….

WHY REMOTE SENSING?

Together with Geographic Information Systems
(GIS) and Global Positioning Systems (GPS),
remotely sensed data can be analysed spatially,
rapidly and accurately for timely decision
making in order to address current issues
relating to
 natural resources security,
 natural disaster control, and
 environmental protection.
 Vegetation?
Different rock types?

Snow depth?

Topography?

HOW TO REMOTELY SENSE …?

 Passive sensors

Object = Target
Eye = Sensor

FOLLOWING OUR ‘DESIGN’

Active sensors

FOLLOWING OUR ‘DESIGN’

 Active sensors

RADAR

LIDAR

TO REMOTELY SENSE
 FORMALLY, A REMOTE SENSING SYSTEM
• A: Source of EMR
• B: Electromagnetic radiation (EMR)
• C: Object of interest
• D: Sensor
• E: Transmission to receiver(s) B
• F: Data products
• G: Results of analyses A

A
 f or v

ELECTROMAGNETIC RADIATION
 • Wavelengths ( ) are the predominant means by which we (Geographic
Remote Sensors) refer to EMR.
• Wavelengths are inversely related to frequency (f) by this equation:
• c= f (where c = speed of light, a constant)
• Thus, as the increases, the f decreases (because c is a constant).
• Frequency is directly related to our ability to detect the wavelength,
since the lower the f, the less the energy
(i.e., photons).

H.W#1 WAVES / PARTICLES

f=v

(think about the relation between temperature and wavelength.)

• As a result, the longer the wavelength, the harder it will be to detect.
This means that
– the detector needs to be more sensitive in order to detect it, and / or
– the ‘source’ area needs to be larger, in order for the sensor to capture
enough energy from the target area.
• This is why, in most remotely sensed data, less spatial detail is available for
passive infrared images.

Far
Infrared
Near
Infrared

Near
Red Infrared
Green

Blue

Sensor Sensor Sensor

WAVES AND PHOTONS

• In remote sensing of the earth, the sensor is looking through a
layer of atmosphere separating the sensor from the Earth's
surface being observed. Hence, it is essential to understand
the effects of atmosphere on the electromagnetic radiation
travelling from the Earth to the sensor through the
atmosphere.

• The atmospheric constituents cause wavelength-dependent

absorption and scattering of radiation. These effects degrade
the quality of images, although some of the atmospheric
effects can be corrected for.

THE ATMOSPHERE
NOT ALL • Absorption is the other main mechanism
GETS at work when electromagnetic radiation
THROUGH interacts with the atmosphere. In contrast
to scattering, this phenomenon causes
molecules in the atmosphere to absorb
energy at various wavelengths. Ozone,
carbon dioxide, and water vapour are the
three main atmospheric constituents
which absorb radiation.
THE DIFFERENCES THAT MAKE A
DIFFERENCE
• In remote sensing of the earth, the sensor is recording the
EMR reflected by the Earth's surface and by the atmosphere.
Generally we are primarily interested in the surface effects,
and therefore understanding how EMR interacts with surface
features is important.

• There will be object-specific wavelength-dependent

absorption, scattering and emission of radiation. These
effects provide the basis for identifying features in remotely
sensed images.

NOT THE ATMOSPHERE

 A surface’s ‘flatness’ is relative
to the wavelength striking it

Examples of RADAR

The differences that make a difference

• The angle at which the EMR strikes the object can have a
significant effect, depending on the surface characteristics.

Topography is, therefore, a significant factor.

THE DIFFERENCES THAT MAKE A

DIFFERENCE
• At any given time our ability to detect EMR is limited by the
current technology.
• Therefore, every sensor reflects a compromise—a decision has
to be made as to which regions to detect (that is, we can’t
detect ALL the useful EMR regions, so we detect those that we
feel would provide the greatest amount of information).

 Every sensor is designed to detect selected

bands within the various EMR regions. Thus,
although spectral reflectance is a continuous
curve, the use of bands discretizes that curve.
Analog versus digital

SENSING A COMPROMISE
 What is the relation between temperature
and EMR wavelength?

The relationship between temperature and EMR wavelength is described

by Wien’s Displacement Law:

As the temperature of an object increases, the peak wavelength of the

emitted electromagnetic radiation decreases (i.e., shifts to shorter
wavelengths).

In short:
Higher temperature = shorter wavelength
Lower temperature = longer wavelength
Geographic Information System GIS

A GIS is a computer-based decision-support tool that allows us

to transform geographic data into geographic information (GI)
GIS allows for the input, storage, analysis and output of
geographic data & information
Input: Aggregate and collate geographic data from
multiple sources
Storage: Retrieve and update data
Analysis: Identify spatial relationships, patterns and
trends across a range of phenomena
Output: Display GI in tabular, graph and map form to
allow other people to interact with the data

Introduction
The Importance of Location

Geographic location is an important attribute of activities,

policies, strategies, and plans

Geographic problems involve an aspect of location, either in

the information used to solve them, or in the solutions
themselves

There are three standard bases for classifying geographic

problems:
a. scale or level of geographic detail
b. timescale ) short, medium, long
c. intent or purpose
problem solving ) normative applications
curiosity driven ) positivist approach

Introduction
The Importance of Location

Introduction
For Clarification
What is the difference between “special” and “geospatial” information??

Geographic ) the Earth’s surface and near-surface

Spatial ) any space (not just the space of the Earth’s

surface)

Geospatial ) a subset of spatial, applied specifically to the

Earth’s surface and near-surface

Introduction
Spatial is Special!

Almost all human activities and decisions involve an important

geographic component
) tied to specific locations on Earth’s surface

It is always multidimensional!

Working with geographic information involves unique,

complex and difficult choices
) which other fields choose not to make

Crucial to understanding both physical and human geography

) understanding the variation in the Earth’s surface
) both natural and man-made

Introduction
GI Systems (GIS) vs GI Science (GISc)

GIS are computer-based tools for collecting, storing,

processing, analyzing, and visualizing GI

GISc is the scientific context/underpinnings of GIS

• CSE (Computer science Engineering)underpinnings
of overlaying map data, faster queries, etc.

• Statistical approaches for analyzing spatial data

GISS combines GI Systems and Science!

Introduction
Data, Information, Knowledge, ...

Information systems enable us to make informed decisions:

Data are numbers or text
) raw facts, a stream of bits

Information comes from refined data, given some amount of

interpretation

Knowledge comes from information only when the information

has been analyzed, understood, and synthesized

The final step is making decisions based on the knowledge

obtained from information, based on data
)Wisdom

Introduction
Decision-Making Support Infrastructure:
List the steps to reach the knowledge by remote sensing.

Introduction
Decision-Making Support Infrastructure:

Introduction
GISS: Six Components
What are the components of GISS?

Hardware: Desktop computers; mobile

devices; cellular phones
Software: Commercial or open source
Data: Digital representation of some
aspect of some specific area
Procedures: Standards to ensure quality,
efficiency, replicability, and accessibility of
GISbased knowledge
People: Expertise in a wide array of
subjects, such as data management,
computer programming, etc.
Network: Rapid communication and
sharing of digital information

Introduction
The Nature of Spatial Variation
What are the principles of the nature of spatial

Three principles of the nature of spatial variation:

1. Proximity e↵ects are key to understanding spatial variation

2. Issues of geographic scale and level of detail are key to

building appropriate representations of the world

3. Di↵erent measures of the world co-vary, and understanding

the nature of co-variation can help us to predict

Representation of Geographic Data

The Task

Representing spatial and temporal phenomena in the real world:

• since the r eal world i s complex, this task i s difficult and
error prone!

• small things (i.e. human lives) are very complex in

detail

• viewed in aggregate, human activity exhibits structure

across geographic spaces

Representation of Geographic Data

The Fundamental Problem

• Deciding what data/information can be discarded as the

inessential while retaining the significant characteristics of the
observable world

• Distinguishing between controlled variation, which oscillates

around a steady state, and uncontrolled variation:

controlled variation ) like utility management

uncontrolled variation ) climate change

Representation of Geographic Data

Tobler’s First Law of Geography

What is Tobler's First Law of Geography?

Everything is related to everything

else, but near things are more
related than distant things

Tobler, 1970

Representation of Geographic Data

Spatial Autocorrelation

Autocorrelation is the similarity between observations as a function

of the time

Spatial autocorrelation is similarity in the location of spatial

objects and their attributes
manifestation of Tobler’s Law!

Is a measure of the degree to which a set of spatial features and

their associated data values tend to be clustered (positive spatial
autocorrelation) or dispersed (negative autocorrelation)

Representation of Geographic Data

Spatial Autocorrelation

how we should represent the real world in our digital GIS

Understanding :
spatial variation, the scale of spatial variation, and the way in
which geographic phenomena co-vary.

Spatial autocorrelation is determined both by similarities in

position, and by similarities in attributes:
positive, zero, or negative

Representation of Geographic Data

Contiguous Spatial Autocorrelation
Moran`s I interpretation

Representation of Geographic Data

Representation

All representation:
are needed to convey information

fit information into a standard form or model

almost always simplify the truth that is being represented

Digital representation:
digital & binary (1s and 0s)

The basis of almost all modern human communication

Representation of Geographic Data

The Fundamental Problem (Again)

Geographic data are built up from atomic elements, or facts about

the geographic world

At its most primitive, an atom of geographic data (strictly, a

datum) links a place, often a time, and some descriptive property

The fundamental problem: “the world is infinitely complex, but

computer systems are finite”

Representation of Geographic Data

What is a Data Model? List the THREE Levels of abstractions that convert reality to data in
the computer.

Levels of abstractions that convert

reality to data in the computer

Conceptual models ) discrete

objects vs continuous fields

Logical models ) rasters vs vectors

Physical models ) Too many

Representation of Geographic Data

Discrete Objects vs Continuous Fields

Discrete Objects ) the world is empty, except where it is

occupied by objects with well-defined boundaries that are instances
of generally recognized categories:
objects can be counted

objects have dimensionality:

0-dimension ) points

1-dimension ) lines

2-dimensions ) areas

Representation of Geographic Data

Discrete Objects vs Continuous Fields

Continuous Field ) a finite number of variables, each one

defined at every possible position:

omnipresent, everywhere dense

can be distinguished by what varies, and how smoothly

In this perspective, value (A) is a function of location (X):

A = f (X )

Contrast with the discrete object view ) define the location of the
boundary of objects, or X = f (A)

Representation of Geographic Data

Rasters vs Vectors

There are two methods that are used to reduce geographic

phenomena to forms that can be coded in computer databases

Each can be used to represent both fields and discrete objects:

usually raster is used to represent fields

and vector for discrete objects

“Raster is faster, but vector is correcter”

Representation of Geographic Data

Raster

In a raster representation geographic space is divided into an array

of cells, each of which is usually square, but sometimes rectangular:

all geographic variation is then expressed by assigning

properties or attributes to these cells

cells are called pixels (short for picture elements)

In the raster data model, individual grid cells have one value that
represent a single phenomenon

IMPORTANT: Raster accuracy is limited by the resolution of the

cell

Representation of Geographic Data

Raster Representation
Each color represents a di↵erent value of a nominal-scale variable
denoting land-cover class

Representation of Geographic Data

Raster Representation
E↵ect of a raster representation using:
(A) the largest share rule
(B) the central point rule

Representation of Geographic Data

Vector

In a vector representation, features are captured as a series of

points or vertices connected by straight lines:

areas are often called polygons

lines are often called polylines

In the vector data model, discrete features can have many di↵erent
attributes representing numerous phenomena

Representation of Geographic Data

Vector Representation
E↵ect of a vector representation:
solid purple line ) represents an area
dashed blue line ) approximation by a polygon

Representation of Geographic Data

Generalization

Simplifying the view of the world:

describe entire areas, attributing uniform characteristics to
them, even when areas are not strictly uniform

identify features on the ground and describe their

characteristics, again assuming them to be uniform

some degree of generalization is almost inevitable in all

geographic data

A geographic database cannot contain a perfect description;

instead, its contents must be carefully selected to fit within the
limited capacity of computer storage devices!

Representation of Geographic Data

Data 6= Information

In the context of computing, both data and information refer to

facts and statistics collected

But data is the quantities, characters, or symbols on which

operations are performed by a computer, and information is the
context or meaning of data

GIS can deal with data easily

GIS can handle information too

But GIS often has trouble to e↵ectively process knowledge and

wisdom

Representation of Geographic Data

Data Types

Data is organized in a GIS database based on whether it is a

collection of text symbols, numbers, or dates
List three classifications for numbers in GIS database.
Numbers are further classified into one of four data types:

Representation of Geographic Data

Data Types
Give an example for a number: precision=5 and scale=4.

Why does this matter? Determine the precision and scale for the number “0.333”.
Determine the precision and scale for the number

because GIS databases can include tens of thousands of

records, it is useful to try and limit the memory allocated to
each record
With numerical values, you can designate:
precision ) the length of the field
scale ) the number of decimal places in the field

Number Precision Scale

13.56 4 2
1056 4 0
0.4326 ? ?

Representation of Geographic Data

GI Attribute Types
List four of the attribute types in

Attribute Types:
nominal ) used to categorize

ordinal ) ordered categories

interval ) di↵erence between values are constant

ratio ) ratios between values are meaningful

Spatially Extensive vs. Spatially Intensive:

extensive ) values aggregated over entire area
intensive ) true for any sub-region of the area

Representation of Geographic Data

 Link the following terms to the one that defines it (make two pairs);
positive spatial autocorrelation, negative spatial autocorrelation,
discrete Objects, continuous Field.

Positive spatial autocorrelation Continuous field

Negative spatial autocorrelation Discrete objects

 Explain ambiguity and vagueness with
an example of each.

Ambiguity: A situation in which the meaning is unclear or open to more

than one interpretation

.Example: "I want a cup of tea with milk

Do you mean a cup of tea with milk added to it
Or do you mean a cup of milk with tea flavor

Vagueness: It means lack of clarity due to lack of details or lack of precise

.boundaries of the meaning of a word or sentence

.Example: The company wants this project as soon as possible

?What does "as soon as possible" mean? Does it mean today? Within a week
‫؟‬Within a month
A Complete Representation of Geography?
It is impossible to make a perfect
representation of the world, so
uncertainty about is inevitable

• Need to identify many of

the sources of geographic
uncertainty and the ways in
which they operate in GI-
based representations

• Also need to be aware of

how uncertainty propagates
through geographical
analysis
Uncertainty

Uncertainty arises from:

the way that GI system users conceive of the world
how they measure and represent it
how they analyze their representations of it

Note that there are a number of conceptual issues in the creation

and management of uncertainty

There are also a number of ways in which it may be measured

using statistical and other methods
Formal Definition?

Uncertainty is a measure of the user’s understanding of the

di↵erence between the contents of a dataset and the
observable phenomena that the data are believed to represent

Not just:
Uncertainty = Reality Representation

But also about how accurately we actually observe that reality, or

are even capable of comprehending reality

A catch-all for incomplete representations or a quality measure

Sources of Uncertainty

• Measurement error:
di↵erent observers
measuring instruments
• Specification error omitted variables

Also includes:
• ambiguity
• vagueness
• quality of a GI representation
Recall Raster Representations
Each color represents a di↵erent value of a nominal-scale variable
denoting land-cover class
Uncertainty in Classification

The rows of the table correspond to the land use class of each
pixel as recorded in the database, and the columns to the class as
recorded in the field

The numbers appearing on the diagonal of the table reflect correct

classification

Classes Conifer Fir Oak Grassland Recorded

Conifer 12 2 0 0 14
Fir 0 13 0 0 13
Oak 0 0 13 0 13
Grassland 0 1 2 13 16
Reality 12 16 15 13 56
Confusion/Error Matrix Statistics

Percent correctly classified (PCC)

total of diagonal entries divided by the grand total, times 100
“overall accuracy”
Here PCC = (51/56) 100 = 91.07%

Problem with PCC

does not reveal if error was evenly distributed between classes

some classes bad, while some really good?
Confusion/Error Matrix Statistics

User’s and producer’s accuracy and types of error

User’s accuracy corresponds to error of commission (inclusion):

ex. 2 Douglas fir pixels included erroneously in conifer
category

Producer’s accuracy corresponds to error of omission (exclusion):

ex. 2 grassland pixels omitted from oak category
User’s Accuracy

From the perspective of the user of the classified map, how accurate
is it?

For a given class, how many of the pixels on the map are actually
what they say they are?

Calculated as:
number correctly identified in a given map class / number
claimed to be in that map class
diagonal value / sum across the row

AccuracyUser ,Grassland = (13/16) 100 = 81.25%

Producer’s Accuracy

From the perspective of the maker of the classified map, how

accurate is it?

For a given class in reference plots, how many of the pixels on the
map are labeled correctly?

Calculated as:
number correctly identified in ref. plots of a given class /
number actually in that reference class
diagonal value / sum across the column

AccuracyUser ,Oak = (13/15) 100 = 86.67%

Assessing Accuracy

The ‘overall’ accuracy of our raster is 91%, but for analysis

purposes, we know that the grassland class is only 81% accurate

Classes Conifer Fir Oak Grassland Recorded User

Conifer 12 2 0 0 14 85.71%
Fir 0 13 0 0 13 100%
Oak 0 0 13 0 13 100%
Grassland 0 1 2 13 16 81.25%
Reality 12 16 15 13 56
Producer 100% 81.25% 86.67% 100% 91.07%

Uncertainty & MAUP

Uncertainty in Attribute Values: Interval/Ratio Case

Here, error is best thought of not as incorrect class assignment,

but as a change of in a value such that the observed value x is
equal to the true value X plus some distortion x

Standard statistical perspective of observation:

observed data = signal + noise

If the average distortion ( ) is zero, with positive and negative

errors balanced out, the observed values are said to be unbiased
Uncertainty in Attribute Values: Interval/Ratio Case

Have to distinguish between:

accuracy the magnitude of the error ( x)
precision
variability among repeated measurements
number of digits used to report a measurement
E↵ects of Geographic Scale

Specifically, the measurement of geographic individuals is most

likely not the intended purpose of particular spatial datasets
areal data not individual data
ex. county poverty rates or census tract income

As a consequence, we cannot assign even dominant characteristics

of areas to individuals or point locations in those areas

This is a significant source of uncertainty in geospatial analysis:

known as the ecological fallacy problem
inappropriate inference from aggregate data about the
characteristics of individuals
E↵ects of Aggregation/Zonation

Choice of zones results in uniform variation in point counts across

areas
Modifiable Areal Unit Problem
I

The e↵ects of scale and aggregation are generally known as the

Modifiable Areal Unit Problem (MAUP)

Since most (if not all) political/administrative boundaries are

arbitrary we can “modify” them to suit our purposes

in the sense that they can be aggregated to form units of

di↵erent sizes or spatial arrangements
Modifiable Areal Unit Problem

The scale problem:

the variation in results that may be obtained when the same
areal data are combined into sets of increasingly larger areal
units of analysis
analysis with di↵erent sample sizes (n)

The zoning problem:

the variation in results due to alternative definitions in the
units of analysis
analysis with the same number of units (n)
Consolidation

Uncertainty is inevitable in GI

Data obtained from others should never be taken as truth

e↵orts should be made to determine quality

E↵ects on GI system outputs are often much greater than expected

there is an automatic tendency to regard outputs from a
computer as the truth

Be honest and informative in reporting results

add plenty of caveats and cautions