[Link].

16 - Physical Principles of Remote Sensing

Book: Satellite Images for Environmental Studies

Chapter 1: IMAGES OBTAINED BY REMOTE SENSING

The sensors installed on artificial satellites are the result of the evolution of
space science and technology. The images obtained from satellites, from airplanes (aerial photographs)
or even on the surface or close to it, such as a photograph of your house,
a school or a landscape of any kind, taken with a common camera, are
all data obtained by remote sensing. For this reason, initially, we will define what it is
remote sensing.

1.1 Remote Sensing

Remote sensing is the technology that allows obtaining images and other types of
data, from the Earth's surface, through the capture and recording of the reflected energy or
issued by the surface. The term sensing refers to the acquisition of data, and remote,
what it means distant, it is used because the acquisition is made at a distance, that is, without the

physical contact between the sensor and the Earth's surface, as illustrated in Fig. 1.1.

Figure 1.1. Acquisition of images through remote sensing.

In Fig. 1.1 we can observe that the sun illuminates the Earth's surface. The energy
coming from the sun, reflected by the surface towards the sensor, is captured and recorded by
this. As we will see later, depending on the type of sensor, the energy emitted by the

1
 [Link].16 - Physical Principles of Remote Sensing

Earth's surface can also be captured and recorded. Note that, in its trajectory, the
energy passes through the atmosphere, which interferes with the final energy recorded by the sensor. How much

the farther the sensor is from the Earth's surface, as is the case with the one onboard
artificial satellites, the greater the interference from the atmosphere. The presence of clouds in

the atmosphere, for example, can prevent the energy reflected by the Earth's surface
arrive at the sensor on board a satellite. In this case, the sensor registers only the energy
coming from its own cloud.
The energy reflected or emitted by the Earth's surface and captured by sensors
electronics, installed in artificial satellites, is transformed into electrical signals, which are
registered and transmitted to reception stations on Earth, equipped with enormous
parabolic antennas (Fig. 1.1). The signals sent to these stations are transformed into
data in the form of graphs, tables or images. Based on the interpretation of this data,
It is possible to obtain information about the Earth's surface. How to interpret images.
obtained through remote sensing you will find in Chapter 4 of this book.

1.2 Energy Sources Used in Remote Sensing

Any activity requires the use of energy, just like obtaining data.
by remote sensing. The energy with which remote sensors operate can be
coming from a natural source, such as sunlight and the heat emitted by the surface of the
Earth, and it can be from an artificial source such as, for example, the flash used in a
camera and the signal produced by a radar.
The energy used in remote sensing is electromagnetic radiation, which
propagates in the form of electromagnetic waves at the speed of light (300,000 km/s). It
It is measured in frequency (in units of hertz - Hz), and wavelength (in
meters). The frequency of a wave is the number of times a wave repeats per
unit of time. Thus, as indicated by Fig. 1.2, the larger the number,
the greater the frequency, and the lower, the lower the wave frequency. The length
wavelength is the distance between two successive wave peaks: the farther apart they are, the greater it is

length and, the less distant they are, the shorter the wavelength will be (Fig. 1.2). A
wave frequency is directly proportional to the speed of propagation and inversely
proportional to the wavelength.
The Electromagnetic Spectrum represents the distribution of radiation

electromagnetic, by regions, according to the wavelength and frequency (Fig. 1.2).

Note that the electromagnetic spectrum ranges from short wavelengths, such as
cosmic rays and gamma rays ) of high frequency, up to long wavelengths
like radio and TV waves of low frequency. In the visible spectrum region, the eye

2
 [Link].16 - Physical Principles of Remote Sensing

humans perceive electromagnetic energy (light), being able to distinguish colors from
violet to red. Infrared radiation (the one from heat) is subdivided into three
regiões: infravermelho próximo (0,7-1,3 m), medium (1.3-6.0 m) it is remote or thermal (6.0-
1000 m).

Figure 1.2. The electromagnetic spectrum.

1.3 The Reflected Energy of the Earth's Surface

The objects on the Earth's surface such as vegetation, water, and soil reflect,
absorb and transmit electromagnetic radiation in proportions that vary with the
wavelength, according to its bio-physical-chemical characteristics. The variation
the energy reflected by objects can be represented through curves, such as those
shown in Fig. 1.3. Due to these variations, it is possible to distinguish the objects from
land surface in remote sensing images. The representation of objects in these
images will vary from white (when they reflect a lot of energy) to black (when they reflect
little energy (Fig. 1.4).

Figure 1.3. Spectral curve of vegetation, water, and soil.

3
 [Link].16 - Physical Principles of Remote Sensing

Analyzing the curves in Fig. 1.3, we observe that in the visible light region the
vegetation (green and healthy) reflects more energy in the range corresponding to green. This fact
explain why the human eye sees vegetation in the color green. However, it is observed that
It is in the near infrared range that vegetation reflects more energy and differentiates itself.
of the other objects. The ground curve indicates a more uniform behavior, that is,
smaller variation of reflected energy, in relation to vegetation, across the spectrum. The water
clean reflects little energy in the visible region and practically no energy in the region
of infrared, while the turbid water (with suspended sediments or polluted)
reflects more energy, but only in the visible region.

Figure 1.4. Image of Ubatuba obtained in the near-infrared range, in channel 4 of

ETM+ sensor (Landsat-7 satellite), 11/08/1999.

In the image of Figure 1.4 we can observe, for example, that the vegetation of
Atlantic Forest, which reflects a lot of energy in this range (as shown in Figure 1.3), is
represented with light shades, while the water, which absorbs a lot of energy in this
band (as indicated in Figure 1.3) is represented with dark shades.
In the visible region, variations result in a visual effect called color.
In this way, a certain object or surface is blue when it reflects blue light and
absorbs the others. The sky, for example, is blue because the air molecules that compose the
the atmosphere reflects in the blue light spectrum. Objects are green, like the vegetation, when
They reflect on the green light band. They are red when reflecting on the red light band.
like the apple, for example, and so on. White light is the sum of the colors of the spectrum
visible, therefore, an object is white as it reflects all colors. Black is the absence of
colors, for this reason an object is black when it absorbs all the colors of that spectrum.
4
 [Link].16 - Physical Principles of Remote Sensing

1.4 Remote Sensors

Remote sensors are devices that capture and record reflected energy.
or emitted by the elements of the Earth's surface. Depending on their characteristics, they
can be installed on land, air platforms (balloons, helicopters, and airplanes) and
orbitals (artificial satellites). The cameras, the video cameras, the radiometers,
The scanning systems (scanners) and radars are examples of sensors.
There are sensors operating in different regions of the electromagnetic spectrum.
Depending on the type, the sensor captures data from one or more regions of the spectrum. The eye

Humans are natural sensors that perceive only visible light or energy. Sensors
artificial devices allow us to obtain data from regions of energy invisible to the human eye.

Figure 1.5. Aircraft BEM-110 BANDEIRANTE (from INPE)

and its main remote sensors. In detail
the camera is installed in the part
the lower part of the plane that flies in stripes, on one side
for another area to be photographed, at intervals
regulars, and triggers the camera
automatically. Photos: Carlos Alberto Steffen.

Cameras capture energy in the visible region and of

near infrared. In cameras, the film acts as the sensor that captures
and registers the energy coming from an object or area. The multiespectral electronic sensor
TM, from the Landsat-5 satellite, for example, is a scanning system that captures data in
different spectral bands (three from the visible region and four from the infrared region).
Radar-type sensors, by generating their own energy source in the region
from microwaves, they can obtain images both during the day and at night and in any
weather condition (including cloudy and rainy weather). This is the main advantage.
two radars in relation to optical sensors that depend on sunlight, such as cameras
5
 [Link].16 - Physical Principles of Remote Sensing

photographic (unless a flash is used), video cameras, multispectral scanners

such as the ETM+ from the Landsat-7 satellite, among others. For these sensors, the
Cloud cover is a limitation in obtaining images. As for synthetic radar,
built by man, the operating principle is the same as that of a natural radar of a
bat. The artificial radar, just like that of the bat, emits a signal of energy to a
object and records the signal that returns from that object.

In fact, the bat relies on the help of a sonar that allows it to capture the
echo of the sounds it emits to locate objects.
Just as it is possible to broadcast a football game in different
radio and TV broadcasters, which operate on different frequencies of energy, called
channels, it is possible to obtain images of the same area in different spectral bands,
also known as channels or bands.
In Figure 1.6, we can observe images of the same area obtained by the sensor
multispectral ETM+ from the Landsat-7 satellite in different channels. By analyzing this figure,
we verified that the objects (water, vegetation, urban area, etc.) are not represented with
the same tone in the different images, because, as we saw earlier (Figure
1.3), the amount of energy reflected by objects varies along the spectrum
electromagnetic and the variations were captured by the ETM+ sensor, which operates in different

channels.

Figure 1.6. Images of Ubatuba, obtained by ETM+ Landsat-7, 11/08/1999, in channel 3

(from the visible region), 4 (from the near infrared) and 5 (from the mid infrared). We can
observe that the urban area is more prominent in the image from channel 3, while the
the separation between land and water is clearer in the image of channel 4. The vegetation is well
dark in the image of channel 3, dark in the image of channel 5 and bright in the image of channel 4,
which, as highlighted earlier, corresponds to the spectral range in which vegetation
reflects more energy.

1.5 Resolution
The resolution refers to the ability of a sensor to "see" or distinguish objects.
from the Earth's surface. More specifically, spatial resolution can be defined as the

6
 [Link].16 - Physical Principles of Remote Sensing

smallest element or surface distinguishable by a sensor. Thus, a sensor such as the

ETM+, which has a spatial resolution of 30 meters, has the ability to distinguish objects that
on the ground, 30 meters or more. This means 30 by 30 meters (900m)2) é
the smaller area that the TM sensor can "see or perceive".
In an aerial photograph or satellite image, with a spatial resolution in
around 1 meter, one can identify the trees of an orchard, the houses, and buildings of a
city or the planes parked at an airport, while in a satellite image,
with a spatial resolution of 30 meters, the orchard will likely be identified,
urban area corresponding to the area occupied by the city and only the airport runway,
as can be seen in Figure 1.7.

Figure 1.7. Images of

San airport
Francisco (USA)
taken with
spatial resolution of
30 meters (a), 5
metros (b) and 1 meter
(c) by the sensors a
board of satellites
Landsat-5, IRS-2 and
IKONOS-2
respectively.
Courtesy: Mauricio B.
Meira.

From the American satellite IKONOS-2, launched in September 1999 (the first
mission, the IKONOS-1, was not successful), it is possible to obtain pancromatic images (region
of visible) high spatial resolution, about one meter, like that of Figure 1.7c, and of four
meters for multispectral images (visible and infrared region).

7
 [Link].16 - Physical Principles of Remote Sensing

1.6 Color Photographs

The tone or color of photographs obtained through a photographic sensor
(camera) will depend on the sensitivity of the film and the filters used in
process of color formation. Thus, with a black and white film
panchromatic, which is sensitive to the visible range, it is possible to obtain black aerial photographs.

and white, also referred to as panchromatic (Figure 1.9a, p.18). with a film
black and white infrared, black and white infrared photographs are obtained,
as simulated in Figure 1.9b.
With a color film, sensitive to the visible range, color photographs are obtained,
also referred to as normals or naturals, in them the objects are represented with the
same colors seen by the human eye (Figures 1.8a and 1.9c). With an infrared film
Colorful, sensitive to the near-infrared range, colorful photographs are obtained.
infrared, also called false color (Figures 1.8b and 1.9d).

Figure 1.8. Natural color aerial photography (a) and infrared color (b) of the University
from Wisconsin (USA). By observing this figure, it is possible to confirm that the soccer field is
made of synthetic grass (f), because if the grass were natural (n) it would be represented
in the red color in photo b, as occurs with the adjacent area of natural grass, and the
remaining natural vegetation.

The colored infrared films were called false-color because the scene,
recorded by this type of film, is not reproduced in its true colors, that is, as
views through the human eye. These films were developed during World War II, with
the goal of detecting camouflaged targets painted green that imitated vegetation.
This detection is possible because vegetation, as indicated by the graph in Figure 1.3, reflects
but more intensely energy in the infrared region. Thus, while in
false-color photographs show vegetation in red, green objects or vegetation
artificial generally appears in blue/green, as can be seen in Figure 1.8b.

8
 [Link].16 - Physical Principles of Remote Sensing

Figure 1.9a. Photographic areas of Florianópolis-SC (in the center, the Ercílio Luz bridges and
Colombo, which connects the island of Santa Catarina to the mainland) - Black and White
panchromatic.

Figure 1.9b. Pictures of the areas of Florianópolis-SC (in the center the Ercílio Luz bridges and
Colombo, which connects the island of Santa Catarina to the Mainland) – Black and White
Infrared (simulation).

9
 [Link].16 - Physical Principles of Remote Sensing

Figure 1.9c. Photographs of areas of Florianópolis-SC (in the center the Ercílio Luz bridges and
Colombo, which connects the island of Santa Catarina to the mainland) – Natural Color.

Figure 1.9d. Photographs of areas in Florianópolis-SC (in the center the Ercílio Luz bridges and
Colombo, which connects the island of Santa Catarina to the mainland) – False Color, in the
we can observe the vegetation represented in red.
10
 [Link].16 - Physical Principles of Remote Sensing
Since vegetation absorbs a lot of energy in the visible spectrum and reflects a lot of energy in the

near infrared, appears dark in (a) and light in (b).

The choice of the type of movie for a specific study will depend on its
the objective and the availability of resources, as colored films are more expensive than those in
black and white. The photographs taken with infrared films are the ones that provide the most
information about vegetation, phytosanitary status of crops (allow for differentiation of plants

sadness of diseased plants) and soil moisture, Figure 1.10.

Figure 1.10. False-color infrared photograph of wheat crops in the Municipality of Tapera–
Note that the areas with healthy wheat are represented in lighter red and
but more uniform, while that of wheat attacked by the disease "foot disease" (Ophiobulus
grass) appears in a darker red mixed with the green that represents the soil.
Courtesy: Maurício A. Moreira.

1.7 Color Images

The images obtained by electronic sensors, in different channels, are
individually produced in black and white. The amount of energy reflected by the
objects will determine their representation in these images in different shades of gray,
between white (when they reflect all energy) and black (when they absorb all
energy). By designing and overlaying these images, through colored filters, blue, green and
red (primary colors), it is possible to generate colored images, as illustrated in the
Figures 1.11, 1.12, and 1.13. In the colored images, the color of an object will depend on the
amount of energy reflected by it, from the mixture of colors (according to the additive process) and

11
 [Link].16 - Physical Principles of Remote Sensing

of the association of colors with images. This association explains the fact that the vegetation and the

urban areas to be represented with different colors in the colored images of the Figures
1.11 and 1.12, although the original images are the same. What changed was just the
association of colors with these images.
Analyzing Figures 1.11 to 1.13, we can verify that if an object is white
in the three black and white images that give rise to the colored image, in this image
(colorful) it is also represented in white as, for example, the sand on the beach
(Figure 1.13). The same process occurs when an object is black in the three images.
originals. That's why it is represented in black as well in the colored image like,
example, the shadow of the relief and the cleaner and deeper water of the ocean.

If an object is light (white) only in one of the original images, in the image
colorful it is represented by the color that was associated with this original image, which explains
the green vegetation in the image of Figure 1.11 and the red vegetation in Figure 1.12. These
were the colors associated with the images from channel 4 of the near-infrared, a region where the
vegetation reflects more energy and appears bright in these images.
If an object appears clear in two of the original images, its color in the image
colorful will be the result of the mixture between the two colors that are associated with
original images in which it is white. Let us take as an example the urban area that
is clearly visible in the images from channels 3 and 5. In Figure 1.11, the images have been associated with

blue and red, respectively. By the additive process of colors, blue mixed
with red results in magenta (pink), which is the color that represents the urban area in
colored image. In the colored image of Figure 1.12, the urban area is represented in
cyan (turquoise) which is the result of mixing blue with green, colors associated
respectively to the images of channels 3 and 5.
These two types of color images (Figures 1.11 and 1.12) are the most commonly used.
In them, the color of objects, in general, is false. Other combinations can be obtained and, among
therefore, we highlight the natural colored image (Figure 1.13), in which the colors of the objects are

true.
From visible Landsat images, for example, it is possible to generate an image.
colored natural, as long as they are associated with their respective colors. Thus, in the example
from Figure 1.12, to the image of channel 1, which corresponds to the blue light band of the visible spectrum,

we associate the color blue with the image of channel 2, which corresponds to the green light band of

visible spectrum, we associate the color green with the image of channel 3, which corresponds to the range of

red light of the visible spectrum, we associate with the color red.

12
 [Link].16 - Physical Principles of Remote Sensing

Figure 1.11.
Colorful image
from Ubatuba,
obtained from
of channels 3, 4 and
5, with the colors
blue, green and
red
respectively.

Figure 1.12.
Colorful image
from Ubatuba,
obtained from
of channels 3, 4 and
5, with the colors
blue, red and
green
respectively.

Figure 1.13.
Colored image
from Ubatuba,
obtained from
of channels 1, 2 and
3, with the colors
blue, green and
red
respectively.

13
 [Link].16 - Physical Principles of Remote Sensing

HISTORY OF REMOTE SENSING

The origin of remote sensing is linked to the emergence of aerial photography.

Thus, the history of Remote Sensing can be divided into two periods: one, from 1860.
one from 1960, based on the use of aerial photographs, and another, from 1960 to the present day,

characterized by a variety of types of photographs and images. The Sensing

Remote is the result of a multidisciplinary effort that integrates advances in Mathematics, Physics,
Chemistry, Biology, Earth Sciences, and Computer Science. The evolution of techniques of
Remote sensing and its application involve an increasing number of people.
from different areas of knowledge.
The history of Remote Sensing is closely linked to military use.
of this technology. The first aerial photograph dates back to 1856 and was taken from a balloon. In

1862, during the American Civil War, the balloon corps of an army was making the
recognition of confederate troops through aerial photographs. Starting from 1909,
the photography taken from airplanes begins and in the First World War its use
intensified. During World War II, there was a great development of
remote sensing. During this period, infrared film was developed, with the
objective of detecting camouflage (mainly to differentiate vegetation from targets
painted green), and new sensors were introduced, such as radar, in addition to occurring
advances in communication systems. Later, during the Cold War,
Several high-resolution sensors have been developed for espionage purposes.
Recently, with the end of the Cold War, many data considered military secrets
have been released for civil use.
In the 1960s, the first orbital photographs (taken from satellites) of
Earth's surface was obtained from the manned satellites Mercury, Gemini, and Apollo.
the most important contribution of these missions was to demonstrate the potential and advantages of

acquisition of orbital images, which encouraged the construction of other collection satellites
from meteorological data and land resources. With the launch of the first satellite
meteorological of the TIROS series, in April 1960, the first records began.
systematic Earth imaging. In July 1972, the first satellite was launched
terrestrial resources, or ERTS-1, later renamed Landsat-1. Currently, in addition to
American earth resource satellites of the Landsat series, there are others such as,
for example, the SPOT series, developed by France. In Brazil, the first images of
Landsat was received in 1973. Today, Brazil receives, among others, images from the satellite.
CBERS, a product of a cooperation program between Brazil and China.

14
 [Link].16 - Physical Principles of Remote Sensing

PROCESS OF COLOR FORMATION

To understand how colored films work, it is necessary to understand the

process of color formation, which can be additive or subtractive. The principle of photography.
colorful consists of the possibility of reproducing any color, from a mixture of
only three primary colors: blue, green, and red. The mixture of the primary colors,
called additive process, forms the colors yellow, cyan (blue-green) and magenta,
What are secondary or subtractive colors. Each of these three colors results from
subtraction of one of the colors of white light. In the additive process of color formation, as
show the diagram, it is observed that the mixture of green light with red light results in
production of yellow light. The mixture of red and blue results in magenta light, and from
mixing green with blue results in cyan light. The combination of the three primary colors, in
equal proportions generate white.
The subtractive color formation process is the most used in generating
fotografias coloridas. Nesse processo, como mostra o diagrama, três filtros são colocados em
in front of a source of white light. The yellow filter absorbs the blue light from the beam of white light and

transmits green and red light. The magenta filter absorbs green light and transmits blue and
red. The cyan filter absorbs the red component and transmits the green and blue. The
superposition of the magenta and cyan filters, shown in the diagram, allows the passage of light
blue, because the magenta filter absorbs green and the cyan absorbs red. The superposition of
yellow and cyan and yellow and magenta generate the colors green and red, respectively.
The superposition of the three filters prevents the passage of light, absorbing the three colors.

primary colors present in white light, and the absence of colors results in black.

15
 [Link].16 - Physical Principles of Remote Sensing

RADAR

The term radar (radio detection and ranging) means target detection and
evaluation of distances by radio waves. Radars operate at wavelengths
much larger than those in the visible and infrared spectral region. They operate in the
microwave region between the K-alpha band (10cm or 40GHz) and P (1m or 300MHz).
The Brazilian territory was mapped at the original scale of 1:400,000 by the system of
GEMS (Goodyear Environmental Monitoring System) radar, transported on board the aircraft.
at 11,000m high. This imaging was carried out in two periods: 1971/1972 and
1975/1976. The first period covered the Legal Amazon, the eastern part of the state of Bahia and
Minas Gerais and northern Espírito Santo; the second period covered the rest of Brazil.
Based on the analysis of these images, a mapping of the natural resources of the entire
Country by the RADAMBRASIL project from 1971 to 1986. The maps resulting from this
projects are published at a scale of 1:1,000,000.
At the orbital level, that is, aboard artificial satellites, civil missions with radar
they began in 1978 with the SEASAT program, developed by NASA. Currently,
notable are the ERS program of the European Space Agency (ESA) and RADARSAT,
developed by Canada, in partnership with NASA and NOAA, of the USA. The ESA has already launched three

Earth observation satellites, ERS-1, ERS-2, and ENVISAT. The ERS satellites, at a
average altitude of 780km, carry aboard a radar that operates in the C band (length of
from the 5.7 cm region of the microwave. The spatial resolution of this sensor is 25 meters.
(size of the smallest object distinguished by the sensor). The ENVISAT, recently launched
(01/03/2002), carries aboard ten sensors aimed at monitoring the use and coverage of the land,
the oceans, polar ice, and the atmosphere. One of these sensors is an advanced system of
radar, the ASAR (Advanced Synthetic Aperture Radar).
The RADARSAT program aims to provide data on sensitive areas of the planet from the point
from an environmental perspective, such as tropical forests, expanding deserts, etc. and for studies in the

areas of geology, geomorphology, oceanography, vegetation, land use and agriculture, among others
others. From this program, RADARSAT-1 was launched, which is at an altitude of 798km. The
Radar, on board this satellite, operates in the C band of the microwave region, with a
spatial resolution that can vary from 10 to 100m.

16
 Book Update
FUNDAMENTALS OF REMOTE SENSING AND METHODOLOGIES OF
APPLICATION

The subject contained in this review aims to complement and/or clarify.

better, some topics about 'sensor systems', included in the book of
Fundamentals of Remote Sensing and Application Methodologies
recently released by me. Thus, only the parts that were transcribed and reviewed
deemed necessary to make it clearer regarding what was said in the book.
The author

SENSOR SYSTEMS

Mauricio Alves Moreira

Senior Researcher
National Institute for Space Research - INPE

INPE
August 2001

1
1. INTRODUCTION

Electromagnetic radiation, when interacting with targets on the Earth's surface, can
to be absorbed, reflected, transmitted, and emitted by them selectively. The proportions of
each of these components of interaction depends on environmental factors and the
inherent to each target. For example, the amount of energy reflected by a plant is
intrinsic characteristic functions of this, but can be altered if the plant is
subjected to stresses such as: nutritional imbalance, water deficit, attack of
pests and diseases, adverse climatic effects, etc. In addition to this, there are the conditions of the
environment, responsible for the attenuation of energy, during the path of radiation
from the target to the location where it is quantified.
Knowing the spectral behavior of the reflected and/or emitted radiation from
different targets on the Earth's surface, it is possible to identify similar targets in
other locations, based only on the similarities of spectral behavior.
On the other hand, if a given target has a spectral behavior
known, any abnormality that occurs in this target is manifested in the quantity of
energy reflected or emitted by it. Therefore, it is reasonable to think that the quantification and analysis
the energy reflected and/or emitted by the targets can be used as a parameter
identifier of each type of target on the Earth's surface.
In light of this knowledge, one might ask. How to quantify this
energy?
There are at least two processes used to quantify this energy: the
contact processes (for example, the thermometers used to measure temperature, the
introduction of intrafoliar sensors, etc) and the processes through remote sensing.
As for the first procedure, that is, direct contact, although it is a method
what is needed is not the interest in remote sensing, considering that the main objective is
the study of targets in large areas. In this case, the second option remains which is to use
instruments that can measure the energy reflected or emitted by targets without touching them.
Thus, it is necessary to build equipment capable of recording the reflected radiation and/or
issued by the targets, one can obtain curves of the quantities of these energies and, thus,
characterize each target in the scene within the sensor's field of view. The curves
The spectra contained in Figure 1 represent the energy reflected by different types of
targets of the Earth's surface and serve to illustrate what was mentioned above.

2
 Fig.1 – Spectral curves of different targets

As can be seen in the figure above, each of the five targets

represented presents, in this range of the electromagnetic spectrum (0.4 – 0.9mm), a
typical spectral behavior which differentiates it from the others. Thus, it is possible to outline
spectral curves of all targets on the Earth's surface, as long as there is a
equipment capable of recording these amounts of energy.

2. Definition

Sensors are devices capable of detecting and recording radiation.

electromagnetic, in a certain range of the electromagnetic spectrum, and generate
information that can be transformed into a product subject to interpretation, whether
whether in the form of images, graphical form or tables, as shown in
Figure 2.

3
 Fig.2 – Products generated from data collected by sensor systems.

3 - CHARACTERIZATION OF SENSOR SYSTEMS

A sensor system is basically composed of a collector, which can be a

set of lens, mirror or antenna, and a recording system (detector) that may be a
movie or other devices and a processor, as illustrated in Figure 3.

Fig. 3 – Component parts of a sensor system.

The sensor systems can be classified according to: the radiation source, to
operation principle and the type of product, as summarized in Figure 4.

4
Fig.4 – Classification of sensor systems regarding category, product type, and source
of radiation.

3.1. As for the Radiation Source

The sensors that measure the radiation reflected and/or emitted by the targets,
come from an external source (e.g. from the sun), are called passive sensors,
because they do not have their own radiation, that is, they depend on a light source
external, as illustrated in Figure 5A.

Fig 5 - Diagram to illustrate a passive sensor (A) and an active sensor (B).

Observe in Figure 5A that the radiation that strikes the targets of the Earth's surface
comes from the sun (external source). This radiation, when interacting with the targets, part of it is

5
reflected, going to hit the detector of a sensor system that is eventually over the
irradiated area. In this category of sensors, we have the SPECTRON-SE590, the Mapper
Thematic (TM) of Landsat, the High Resolution Visible (HRV) of SPOT, etc.
On the other hand, if the sensor system has a source of radiation, that is, it does not
depends on an external source to irradiate the target, it is said to be active. In this case, the sensor
emits a radiation flow in a specific spectral range that interacts with the targets
on the surface of the Earth and the part that is reflected is then captured by the sensor. As
Examples of active systems include radars, lasers, and microwave radiometers.
cameras when using "flash" as a radiation source (Figure 5B).

3.2. Regarding the Operating Principle

Passive and active sensors are categorized into two groups regarding
operating principle: scanning and non-scanning
scanning.
Non-scanning sensor systems register the reflected radiation from an area.
of the Earth's surface in its entirety at the same moment. For this reason, they are
also known as frame sensors ("freming systems"). Radiation data
collected by these sensors can be expressed in two forms: image and non-
image (graphic or numeric). In this category, we have photographic systems (form
(image) and radiometers (do not form an image).
On the other hand, in scanning systems, the image
the scene is formed by the sequential acquisition of 'elementary images of the terrain' or
"resolution element", also known as "pixel" (Novo, 1992). According to
exemplified in Figure 4, we have the electro-optomechanical sensors, the cameras
CCD, the radars, among others.
In the literature, other ways to classify systems can be found.
sensores. Entretanto, todas elas têm como base à fonte de radiação e o produto gerado.
For example, Novo (1992) adopts the classification of sensor systems into two
categories: image intensifiers and non-image intensifiers, considering the radiation source
(liabilities and assets). The category of non-imaging systems includes
radiometers and, in the category of imagers, photographic systems, sensors of
electro-optical-mechanical scanning, side-looking radars, etc.

3.3 - Regarding the Type of Product

As for the type of product, the systems are classified into photographic systems.
and non-photographic. In this case, the non-photographic systems include the non
image creators and image creators.
The idea of classifying sensors into photographic and non-photographic categories is
specifically to distinguish the radiation registration process and the form of their product
generated. For example, while in photographic products energy is represented by
continuous manner (analog), in the products generated from non-photographic sensors
energy is represented in discrete amounts.
Photographic systems are devices that, through an optical system
(lens set), record the energy reflected by the targets on the Earth's surface in
a photosensitive film, which are photographic films (detectors). For the purposes
In remote sensing, the most commonly used photographic systems are those
aerially transported, among them we can mention the metric cameras.

6
 Non-photographic sensors are devices used to measure intensity of
electromagnetic radiation, coming from the targets of a surface of the Earth, in
certain spectral bands of the electromagnetic spectrum. The equipment that
Devices that have non-photographic sensor systems are called radiometers.
On the other hand, radiometers are passive systems that measure the intensity of
radiant energy (radiance), coming from all points of a surface, within
of the field of view (FOV), in certain spectral regions of the spectrum
electromagnetic. These regions are usually selected through a filter that
blocks the remaining radiation that strikes the system. The radiometers can be
classified as non-imagers and imagers.

3.3.1-Non-imaging radiometers

Non-imaging radiometers measure the target's radiance and present the

data in the form of graphs or numerical.
When a radiometer expresses the data in numerical form, it is said
band radiometer. In this case, the equipment registers the radiation coming from
targets in broad bands of the electromagnetic spectrum. As an example, we could cite the
Exotech radiometer, which operates in 4 spectral bands (0.5mm – 0.6mm; 0.6mm – 0.7
mm; 0.7mm – 0.8mm; 0.8 – 1.1mm). The value of the radiance generated in each of these
bands or spectral bands, corresponds to an average value of radiance in each
wavelength contained within these ranges or spectral bands.
It is worth noting that there are band radiometers that operate only in a
given spectral range, as is the case with the thermopoint (operating from 8mm to 14mm), used
to measure the temperature of targets or the quantum line sensor (LAI – 2000) that operates
in the spectral band from 400 nm to 700 nm, used to measure radiation
photosynthetically active. Most of these sensors operate in the spectral regions of
thermal infrared (8.0mm to 14.0mm) and microwaves (0.3mm to 30mm).
The radiometers, whose data allows to obtain a curve of the radiance values,
in a certain range of the electromagnetic spectrum, are referred to as
spectroradiometers. As an example of a spectroradiometer, we have the SPECTRON
– SE590 that provides 255 values of radiance from the target, in the spectral range
between 400 nm and 1100 nm. Thus, while in the first case some values are obtained from the
radiance or the energy emitted, in the second it is possible to obtain a much larger number
of these values that allows to draw a curve representing the energy reflected by the
targets.

3.3.2-Imaging radiometers

Imaging radiometers are equipment whose final results are in

form of images of an area (scene) on the terrain. In these devices, the element of
land resolution (ERT) moves generating a sweeping line. Under the
the designation of imagers (“scanners”) there is a wide variety of sensor systems
non-photographic, which record electromagnetic radiation in different ranges of
electromagnetic spectrum, from ultraviolet to far infrared
(approximately from 0.2mm to 14mm).
The characteristics of non-photographic imaging sensor systems in general
are expressed in terms of four resolution domains, namely: spectral, spatial
geometric, temporal-radiometric, which we will discuss next:

7
Spectral resolution: refers to the resolving power that the sensor has to discriminate
different targets on the Earth's surface. In other words, it refers to the best or the
worse characterization of the targets as a function of the spectral bandwidth in which the sensor
The thinner the bandwidth that a given sensor operates, the better.
is its spectral resolution. On the other hand, if a sensor system has detectors
operating in more than one spectral band of the electromagnetic spectrum, the system is said to
multispectral, because it records the electromagnetic radiation coming from the targets in
various spectral bands, for example, the sensor systems onboard satellites
Landsat which we will talk about later.
Sometimes, within the sensor system itself, there are different spectral resolutions.
For example, the MSS sensor of Landsat operates in four spectral bands, that is, MSS-4.
(500 - 600 nm), MSS-5 (600 - 700 nm), MSS-6 (700 - 800 nm) and MSS-7 (800 - 1100
The spectral resolution of bands MSS4, 5, and 6 is 100 nm, while that of the band
MSS-7 is 300 nm.

Spatial or geometric resolution: refers to the instantaneous field of view (IFOV,

Instantaneous Field of View
The spatial resolution of an imaging system is not an easy concept to
define (Mather, 1987). According to the author, it can be evaluated in numerous ways,
depending on the user's goals. Townshend (1980) uses some criteria
based on the geometric properties of the imaging system to define the
{"spatial_resolution":{"a":"ability to distinguish targets from one another;","b":"ability to measure"}}
the periodicity of repetitive targets and the ability to measure spectral properties
of small targets.
The IFOV corresponds to the area on the ground that is 'seen' by the system.
sensor at a given altitude at a given moment in time. The IFOV can be measured from
two ways: in relation to the angle of incidence or in relation to a distance in
land (X,Y), as shown in Figure 6A. In fact, the distance X,Y
corresponds to the diagonal of the circle (Mather, 1987).

Fig. 6 – Diagram to represent the IFOV and PSF

8
 IFOV is the most frequently cited measure of resolution, although it is not
necessarily the most practical. The source of radiance (point on the ground) is 'seen' over
an image like a diffuse circular region. This fact makes the energy distribution
of the Gaussian type, as shown in Figure 6B. In this case, the energy depends
the optical properties of the system components and the relative brightness
from the radiation source (target). This distribution is referred to as the Function of
Diffusion of the Point (PSF). Therefore, the definition of spatial resolution as a function of IFOV
it is therefore not completely satisfactory and is based solely on the resolution
geometric, not taking into account the optical properties of the source (target).
However, it is the most used and it is what was adopted in this document to define resolution.
geometric or spatial.
The IFOV or the PSF should not be confused with the size of the 'pixel'. One
Digital image, like satellite data, is an ordered set of values.
numeric, where each value is related to the radiance of an area on the ground
represented by a simple cell or "pixel". The dimension of the "pixel" does not
it is necessarily related to the IFOV. To make what was said clearer,
let's take as an example what happens in the MSS sensor (Multispectral Scanner)
Subsystem). The energy reflected from the Earth's surface is directed by the mirror
rotating to an optical fiber, which transmits this energy to a series of filters.
These filters separate the signal arriving into four spectral bands. Each of the 4
MSS bands have a set of six detectors which generate analog signals
proportional to the energy that arrives. These analog signals are sampled every 9.958
ms, corresponding to 56 meters on the ground. Since the IFOV of the MSS is
normally 79 m, this means that a sample of this signal (sampled) is
related to an area of 79 m in the direction of the satellite's movement and 56 m in the
transversal direction (size related to 9.958ms). Thus, we will have an area of
56 m in the sensor scanning direction and 79 m in the direction of displacement
satellite. This pixel size (56m x 79m) is smaller than the nominal size of
IFOV. For the pixel to be 79 m x 79 m, as is the size of the pixel
in MSS, there is a sampling of about 40% in the transverse direction. After there was
the sampling, the signal is converted from analog to digital (discrete values) with a
radiometric resolution of 6 bits (64 levels) and is transmitted to stations of
tracking in the soil.
What would this improvement in spatial resolution imply for orbital data, that is,
place satellites with sensors with smaller IFOV?
Let's imagine the IFOV of the MSS (80m x 80m = 6,400 m). If we2 have three in that area
distinct targets (forest, grass, and crop), in the data collected by MSS the radiance
resulting from these targets (average radiance) will result in a single signal,
consequently, the targets cannot be individualized in the MSS images. For
On the other hand, this same area to be covered by the TM sensor (Thematic Mapper,
IFOV = 30m x 30m or 900m2would require almost 9 'pixels' from the TM. This means
that the targets in the MSS data are not individualized, now with the data from
It is possible to distinguish three values of radiance in the area of 6,400m.2Logo, for certain
objectives, the use of TM data is better than that of MSS. Figure 7 shows
how can this be possible.
It is worth emphasizing that the edge pixels, that is, located between two targets
different, in the case of forest and culture or pasture and culture or even pasture and forest, continue
being pixels that contain a mixture of energy reflected by the two targets. For these
The pixels' radiance (reflected energy) will be represented by the dominant targets. In our

9
 %RRN8SGDWH
)81'$0(17$/6 2) 5(027( 6(16,1* $1' 0(7+2'2/2*,(6 2)
$33/,&$7,21

6(1625 6<67(06

0DXULFLR
$OYHV
0RUHLUD
6HQLRU
5HVHDUFKHU
1DWLRQDO
,QVWLWXWH
IRU6SDFH5HVHDUFK
,13(

,13(
$XJXVW
 For airborne imaging systems, the IFOV usually ranges from 1 to
2.5 milliradians (mrad), that is, from a height of 1,000 meters the ground resolution is 1
a 2.5 m (Chipman, 1990), while for the Landsat MSS sensors the resolution is
about 0.083 milliradians, which is equivalent to a ground resolution of
2
approximately 79 m x 79 m or an area of 6,241 m. The TM sensor has a
spatial resolution of 0.032 mrad which results in a ground resolution of 30 m x 30 m
(area of 900 m2This, from the perspective of mapping agricultural areas by
example, it means that the area calculation follows the following order of precision: sensors
aerotransported > TM sensor > MSS sensor.

Temporal resolution: this resolution is a function of the characteristics of the platform on which
the sensor is installed. In the case of orbital sensor systems, the temporal resolution
indicate the time interval that the satellite takes to cover the area of interest again.
This depends on the width of the imaged strip on the ground. For example, the TM sensor of
Landsat-5 has a temporal resolution of 16 days, meaning that every 16 days the Landsat-5
passes over the same geographic point on Earth. Thus, we say that the resolution
The temporal resolution of the TM sensor is 16 days. The sensor systems on board the NOAA satellite
they have a temporal resolution of 9 days, however, as the bandwidth is very
great, it is possible to obtain daily data on the same point.
Temporal resolution is very important because it allows for a ...
dynamic monitoring of targets on the surface of the Earth.

Radiometric Resolution: the electromagnetic radiation, reflected and/or emitted by the targets
from the Earth's surface, has an intensity value that differs from one target to another. For
For certain wavelengths, vegetation reflects and/or emits a lot.
less energy than a zinc plate. However, certain targets, despite being
different, reflect or emit electromagnetic radiation with intensity values
very close to each other, becoming almost identical spectrally. Thus, the
The radiometric resolution of a sensor refers to the sensor's ability to
being able to discriminate, in an imaged area, targets with small differences in radiation
reflected and/or emitted.
On the other hand, the radiation striking the detector is transformed into an electrical signal,
amplified and retransmitted to a receiving system or recorded on devices
(recorders), contained in the platform. Thus, it can be said that the resolution
Radiometric is associated with the ability of the sensor system to discriminate signals.
electric with small differences in intensity.
To represent these values of the intensity of the electrical signal, which are sent
and recorded on board, shades of gray are assigned, which range from dark to the value of
intensity equal to or close to zero, and light gray for the highest maximum value
intensity. Between these two extremes, shades of gray are associated that range from dark
the light, generally referred to as light gray, medium gray, medium gray-
dark, etc.
The number of shades of gray that a given satellite product can present
it depends on the sensor's recording system. For example, if the signals are recorded in 2
bits, the radiometric resolution will be 4 digital levels or gray levels. This means
that, in a given area imaged by this sensor, has 50 targets with reflectances
different, these reflectances will be represented by only 4 levels of gray. It is possible to
to realize that many of these targets will be grouped into a single class, that is, there will be
a grouping of targets whose reflectance values will be close to each other, due to
to the small variation in gray level. To calculate the radiometric resolution, the following is employed

11
the expression2n ,
sendo “n” o número de bits utilizados na gravação. O exemplo a seguir
illustrates well what was discussed above. Imagine two sensors A and B. Sensor A records the
electrical signals in 4 bits and sensor B in 5 bits. To know the radiometric resolution of
each one employs the expression 2nThus, sensor A has a radiometric resolution
of 16 digital levels (24= 16), while that of sensor B will be 32 (25= 32). This means
to say that sensor A is only able to separate the entire range of values into 16 levels
intensity of the radiation reflected and/or emitted by the targets in the imaged area. The sensor
B can individualize 32 levels. Therefore, the radiometric resolution of sensor B is
better than that of sensor A. In Figure 9, two images of Brasília can be seen with
two and four shades of gray.

Fig.9 - Simulation of a Landsat-TM image with two and four levels of gray.
obtained about the city of Brasília.

In some imaging systems, the signal from a detector modulates the

the flicker of a lamp whose light impresses a photographic film, according to lines
perpendiculares aocomprimento do mesmo. Desta forma, obtém-se uma imagem muito
similar to a conventional photograph. However, in most systems
image recorders, the signal from the detectors is recorded on magnetic tapes, or in
in the case of unmanned satellites, it is transmitted to ground stations, as an example, the
data reception station for satellites, located in Cuiabá - MT.

4. LEVELS OF SPECTRAL DATA COLLECTION

The procedure for acquiring the measurements of the spectral properties of the targets
from the Earth's surface, through sensor systems, can be done at three levels:
terrestrial, suborbital, and orbital.
At ground level, sensor systems can also be installed on masts,
placed on boats, fixed on buoys or secured inside laboratories. At the level
Suborbital, generally, aircraft is used as a data collection platform, and for
at the orbital level, unmanned satellites and balloons are used, as can be seen in the
Figure 10. There are also sensor systems that are used in orbital platforms.
manned, but for more specific research use.

12
Fig. 10 – Levels of collection of the energy reflected and/or emitted by the targets on the surface of the
Earth.

The use of a given sensor or another, at a certain level of data collection

spectral information depends mainly on factors related to: a) objective of
research; b) size of the imaged area; c) availability of sensor equipment and
d) cost and desired accuracy of the obtained results. However, some sensor systems
are more extensively used at certain levels of data collection than
others. For example, portable radiometers and spectroradiometers are
equipment widely used to obtain spectral information in areas
experimental. The photographic sensors (photogrammetric cameras) and, more
recently, hyperspectral radiometers are airborne because they are
equipment configured to operate on this type of platform, that is, at level
airborne. The 'scanners' (imagers) are widely used in unmanned satellites.
like in Landsat, in SPOT, among others.

5. SENSORS USED FOR DATA COLLECTION AT LEVEL

ORBITAL

A idéia de desenvolver um sistema sensor, para registrar a radiação refletida e/ou

emitted by the targets on the Earth's surface, from orbital platforms, arose after the
analysis of the first photographs taken of the Earth, during the Mercury programs
Gemini. To that end, these sensor systems should be capable of producing images
snapshots of the terrain, in a manner similar to the photographic systems that were
used in the Mercury and Gemini programs. In this way, the RBV system was designed.
(Retum Bean Vidicon), similar to a television camera that allowed for the recording
snapshot of a certain area of the land. The RBV operated on the first two satellites of
Landsat series, in three spectral bands and, in Landsat-3, with only one channel

13
panchromatic, with a spatial resolution of 40m x 40m (pixel). Table 1 contains
some characteristics of the RBV system, which was part of the payloads of the satellites
Landsat-1 and 2, (Novo, 1989).
In the RBV, the energy coming from the imaged area caused an electric stimulus.
on a photosensitive surface of the camera tube and, for a certain time, the input
of energy was interrupted by a shutter, so that the image of the terrain that had
the image was scanned by a beam of electrons. The video signal was then
transmitted by telemetry.

TABLE 1 – CHARACTERISTICS OF THE RBV SYSTEM

CHANNEL STRIP REGION SPECIAL RESOLUTION
SPECTRAL(nm) SPECTRAL (mxm)
1 475 – 575 BLUE-GREEN 80
2 580 – 680 GREEN
RED 80
3 690 – 830 RED
IV/NEXT 80

After the conception of the RBV, many other orbital sensor systems were
developed for the purpose of collecting information on natural resources. We will discuss the
follow the operating principles of some of these sensor systems more
known in our midst.

5.1. Electro-Opto-Mechanical Imageators

The electro-optomechanical imagers are basically made up of a

set of rotating mirrors or prisms, to collect electromagnetic radiation
coming from the imaged surface targets; an optical system capable of focusing the
energy collected on detectors, which converts it into an electrical signal (Figure 11).

Fig.11 - Diagram of an electro-optical-mechanical sensor system

SOURCE: adapted from FAO (1989).

14
Operating principle: the electromagnetic radiation coming from the targets
contents within the land area imaged by the sensor are collected, initially, by the
rotating mirror E1 (driven by motor M) and reflected to a second mirror
parabolic E2. In the parabolic mirror, the radiation undergoes a new reflection reaching the
E3 mirror, which is a type of filter that separates radiation into two components.
the first component consists of short wave radiations that are transmitted through
of the mirror going to the detector D1. The second component corresponds to the radiations
of long waves that are reflected to the detector D2. The detectors are coupled under
containers containing liquid nitrogen to cool the system. Two black bodies
CN1 and CN2 are placed close to the rotating mirrors, which emit radiations.
about them for system calibration.
Electro-optomechanical imagers are systems with good resolution.
spectral and low geometric or spatial resolution. This category includes, among
others, the MSS (Multispectral Scanner Subsystem) imager and the TM (Thematic
Mapper), both components of the payload of the Landsat series satellites.

5.1.1 – MSS Sensor System

In fact, the MSS system was conceived at the same time as the development of
RBV system. The RBV and the MSS were the first sensor systems deployed in
orbit, in the year 1972, with the aim of collecting data on natural resources
renewable and non-renewable resources of the Earth's surface. Unlike the RBV, which was part of
only in the first three satellites of the Landsat series, the MSS system was and still is one of the
sensors constituting the payload of all satellites in the series launched to date
Landsat 1, 2, 3, 4, 5, 6, and 7.
The MSS system consists of an oscillating mirror, an optical part and a
detection system that records the electromagnetic radiation reflected by the targets of
Earth's surface, in four spectral bands. For each spectral band there is
6 detectors, thus allowing imaging of 6 lines on the ground at the same time.
The detectors produce a voltage (between 0 and 5 volts), which is proportional to
amount of radiation that reaches them, coming from the targets contained in the lines
The voltage produced is an analog signal that is converted into values.
digital (quantized) or gray levels ranging from 0 to 63. This conversion represents the
radiance or brightness of each target in the imaged area. In Figure 12, one can see a
basic configuration diagram of the MSS sensor system.

15
 Fig. 12 - Diagram of the MSS imager and its constituent parts.
[Link]

In the first three satellites (Landsat 1, 2, and 3), the MSS system was placed in
orbit at an altitude of 920 km. To make one orbit around the Earth, the satellite
spent 103 minutes and 27 seconds, totaling, at the end of a day, 14 imaged tracks
with a width of 185 km. The satellites were rotating in a synchronous orbit with the sun with a
inclination angle with respect to the equator of 99º11', causing the orbit to be
almost polar around the Earth.
The orbit configuration of the first three satellites was established in such a way
that every 18 days, the MSS system imaged the same region of the Earth's surface,
that had been imaged 18 days ago, thus assigning a temporal resolution to it
18-day system. In Figure 13, the platform of the first three satellites can be seen.
of the Landsat series and a diagram of the satellite with its main components. In
Table 2 contains information about the width of the spectral band, region of the spectrum
the spatial resolution of each set of detectors that make up the MSS system.

16
 Fig. 13 – Photo of the platform of the Landsat 1, 2, and 3 satellites.
SOURCE: EOSAT (1992).

TABLE 2 – SPECTRAL AND SPATIAL CHARACTERISTICS OF THE MSS SENSOR

Band SPECTRAL RANGE REGION OF RESOLUTION
SPECTRUM SPACIAL
nm mm (mxm)
4 500 – 600 0.5 - 0.6 Green 80
5 600 - 700 0.6 - 0.7 Red 80
6 700 – 800 0.7 - 0.8 IVNext 80
7 800 – 1100 0.8 - 1.1 Next 80

The detectors used in the MSS are photomultiplier tubes for bands 4, 5.
and 6 e, for the band 7, the silicon diode metal alloy is used. In Landsat-3, the
the MSS sensor was composed of another set of detectors that operated in the range
thermal infrared spectral (10.4 to 12.6mm) with spatial resolution of 237m x
237 m.

5.1.2 – Imageador TM

Starting in 1984, NASA launched the TM (Thematic) sensor into orbit.

Mapper) with the same characteristics as the MSS, but with a series of improvements in
concerning the resolutions: spectral, spatial, temporal, and radiometric. This
system, just like the MSS, became part of the payload of the Landsat-4 satellites,
5, 6 and 7, with the Landsat-6 system not having operated for reason of
destruction of the satellite during launch. Figure 14 contains the configuration
from the platform used on Landsat 4, highlighting the positioning of the sensor
TM, whose operating scheme is shown in Figure 15.
Payload is understood as those equipment that is onboard.
a satellite exclusively for collecting information about the Earth's surface
(New, 1989).

17
 Fig. 14 – Photo of the Landsat 4 satellite with detail for the TM sensor

Fig. 15 - Thematic Mapper Sensor System - TM

The TM sensor operates in 7 spectral bands of the electromagnetic spectrum being: 3

bands in the visible region, 3 bands in the reflected infrared region and 1 band in the
thermal region. The width of the imaged strip is also 185 km. The spatial resolution,

18
para os sensores que operam nas regiões do visível e infravermelho refletido, é de 30 m
x 30 m e, for the sensor in the thermal region is 120 m x 120 m (Table 3). In Figure
16 shows part of an area in the State of Rio de Janeiro, imaged in the seven bands
spectral bands of the Landsat-5 TM
On Landsat-7, launched in 1999, a detector was added that operates in a
panchromatic band (520 to 900 nm), with a spatial resolution of 15m x 15m. For the
the other spectral bands there were some modifications as shown in the
Table 4 and Figure 17 and 18.

TABLE 3 - SPECTRAL AND SPATIAL CHARACTERISTICS OF THE TM SENSOR

Band SPECTRAL BAND REGION OF RESOLUTION
SPECTRUM SPACIAL
Nm mm (mxm)
1 450 - 520 0.45 - 0.52 Blue 30
2 520 – 600 0.52 - 0.60 Green 30
3 630 - 690 0.63 – 0.69 Red 30
4 760 – 900 0.76 - 0.90 IV/Next 30
5 1550 – 1750 1.55 - 1.75 IV/Medium 30
6 10400–12500 10.4 - 12.5 IV/Thermal 120
7 2080 – 2350 2.08 - 2.35 IV/Medium 30

Fig. 16 – Images of an area in the State of Rio de Janeiro, obtained in the seven bands of
sensor TM of Landsat-5.

19
TABLE 4 - SPECTRAL AND SPATIAL CHARACTERISTICS OF THE SENSOR
ETM+ OF LANDSAT-7
Band SPECTRAL RANGE REGION OF RESOLUTION
SPECTRUM SPACIAL
Nm mm (mxm)
1 450 – 520 0.45 - 0.52 Blue 30
2 530 - 610 0.53 - 0.61 Green 30
3 630 - 690 0.63 - 0.69 Red 30
4 780 – 900 0.76 - 0.90 IV/Next 30
5 1550 – 1750 1,55 – 1,75 IV/Medium 30
6 10400– 12500 10.4 - 12.5 IV/Thermal 120
7 2090 - 2350 2.08 - 2.35 IV/Medium 30
8 (PAN) 520 - 900 0.52 – 0.90 VIS/IVNext 15

Fig. 17 – Sensor ETM+

[Link]

20
 Fig. 18 – Operation diagram of the ETM+ sensor.
SOURCE: [Link]

It can be observed in Figure 16 that the intensity of the radiance coming from the targets
Imageados differ from one band to another. This change in the intensity of the radiance is
function of the interaction of electromagnetic radiation, in the considered range, with each target
within the imaged area.
In the TM6 band, the energy collected by the sensor comes from emitted radiation.
gray levels of the targets in this image
represents different surface temperatures. In other cases, the gradation of levels
of gray refers to the greater or lesser energy reflected by the targets in the spectral range that
operates each sensor.
While the MSS sensor has 6 detectors per band, the TM each band
The spectral range of visible and reflected infrared is composed of 16 detectors.
For the thermal band, the sensor is composed of a set of 4 detectors. In addition
In the TM, the radiance that reaches the detectors is quantified into 256 digital numbers.
or gray levels, compared to 64 levels in the MSS. The temporal resolution was also
melhorada em relação ao sensor MSS, passando de 18 dias (sensor MSS) para 16 dias.

The other passive and active sensor systems follow

as it is in the book

21
 Bibliography and consulted websites
Chipmam, R. International cooperation in the acquisition and dissemination of
satellite remote sensing data. In: FAO – Food and Agriculture
Organization of the United Nations. Rome, 1990. p 11-26
FAO – Food and Agriculture Organization of the United Nations. Remote
sensing applications to land resources. In: Proceedings. 14th UN/FAO
International Training Course in cooperation with the Government of Italy
Rome, Italy. 1989. 299p.
[Link]

[Link]

[Link]
_abs.html

[Link]

New, E.M.L.M. Remote Sensing: Principles and Applications. They are

Paulo. Ed. Edgard Blucher, 1989. 308p
New, E.M.L.M. Remote Sensing: Principles and Applications. They are
Paulo. Ed. Edgard Blucher, 2nd edition, 1992. 308p
Mather, P.M. Computer processing of remotely-sensed images: An
introduction. New York, Ed. John Wiley & Sons. 1987. 352p.
Townshend, J.R.G. The spatial resolving power of Earth resource
satellites: a review. NASA Technical Memorandum 82020. Goddard
Spaceflight Center, Greenbelt, Maryland. 1980.

22
Editing and creation oftopics and tables based on nonexistent data.

At this point, we will address the tools for creating and editing geographical data available.
in the program. This is a very sensitive aspect for those who work with GIS, given that this is the
phase that consumes the most resources. It must also be remembered that the entire process of
spatial decision will directly depend on the data used: wrong data can
lead to wrong decisions.

We can create geographic data if it contains location information such as e.g:

addresses (address geocoding) or x and y coordinates collected with GPS receivers
or Satellite Images (this method will be addressed later). Another way to introduce
data in a GIS in vector format is through digitization done on the screen (on-screen
digitizing) from aerial photographs or satellite images and the other is Through the filtration of a
theme (Query Builder).

1.1. Filtering a theme (Query Builder).

A group of entities can be selected based on their attributes using the

QB tool. This tool can be used as a filter to establish
viewing parameters of entities that meet a desired criterion. To do this, they use
conditional operators ( < less than, > greater than, = equal to, ≤ less than or equal to, ≥ greater than or equal to

same).

As an example, if we want to identify which is the driest district in Mozambique?

We can base ourselves on the annual average precipitation data of Mozambique, or the district with
precipitation less than or equal to 350 mm.

For this, use the data on precipitation in Mozambique, and open in the Climatology folder, add the
precipitation map. Open the attribute table and view the value window, if it exists.

Then, look for the QB tool by double-clicking on the Precipitation theme, and on the Layer.
property solicitationDefinition Query>Query Builder and insert the conditional operation (≤). This
Value ≤ 350.
Figure 1: Use of the QB tool
1.2. Digitalization.

Many times we have to convert data that is in another format, for example CAD to
Shapefile or geodatabase so we can edit in ArcGis.

It is possible to connect ArcMap to a digitizing tablet and from there vectorize the various
elements that make up the map. However, it is important to mention that this data acquisition method
Vectorization will not be addressed in this Unit of Study.

Based on vectorization from the screen, we will digitize some of the buildings that
found in the image. It is important to mention that based on this procedure it is possible to digitize
objects in vector or raster formats and obtain point, line, polygon.

Let's digitize the buildings and CDS area. To start, let's initialize with ArcMap.
Add the image of CDS taken from Google Earth, which is located in the folder, to the desktop.
database > raster > CDS.

Then activate Editor, from the customize tool, that is Customize > Toolbars > Editor.
A new data editing tool will appear in the ArcMap toolbar.

Figure 2: Activating the Editor in ArcMap 10

After activating the Editor, enter ArcCatalog and create a New Shapefile (Fig 10).
Figure 3: Creation of a new shapefile in ArcCatalog

When creating a new shapefile, describe the characteristics of the objects to be digitized, for the case
Specifically, we want to digitize buildings that are in polygon form.

Figure 4: Creation of a new shapefile

Click ok, and the created Shapefile will automatically appear in the table of contents, at the beginning.
from digitization, Click on the editor > Start Editing, the Create Feature window will appear. Click on the
theme buildings >Polygon.

Figure 5: Digitization of buildings

Then start the digitization of each building (Fig 5). However, it is important to note that for
each object must create a new shapefile, otherwise there will be an overlap of the
data.
Figure 6: Digitization of buildings

After the digitization of Buildings, finish the digitization by clicking on the stop Editing button.
editor window. Then edit the attributes table, so that each building can be assigned.
its respective name, as already mentioned, each theme is associated with a table of attributes.

For the attribute table editing, open the attribute table, add a new column for
insert the names of the objects or building and assign the name ofNAME. and edit the table so that it
can write the names of objects or buildings. After this procedure we write the
names of the buildings, the case of the UniZambeze offices, and the respective rooms, office of
CDS, guesthouses, and basketball field and we finished the map editing.
Figure 7: attribute tables

Figure 8: Identification of names.

1.3. Create themes from coordinate files.

The acquisition of data with GPS prescribes its subsequent visualization, and this can be done at
starting from ArcMap, however, two procedures are basic for visualization:
Download the GPS data from a specific program, such as DNGarmin, and
posterior visualization.
Based on an Excel file ([Link]), this procedure is based on
XY coordinates of the data collected with GPS. With the aim of visualizing the
dados no ArcMap, Abra o ArcMap. Seleccione a opçãonew empty map e click OK
Then add the file [Link] to ArcMap by right-clicking on the
table name select the option Display XY data (alternatively, use the menu
Tools > add XY data).

2. BIBLIOGRAPHY

ROCHA, C. H. B. Geoprocessamento: tecnologia transdisciplinar. Juiz de Fora: s.n., 2000.

SILVA, J. X. da. Geoprocessing for environmental analysis. Rio de Janeiro: Author's Edition,
2001
Gomes & Aguais, Basic Notions on Geoprocessing: UFG, 2005