2/11/2016

Photogrammetric Surveying

Content
 Introduction
 Basic principle
 Elevation of a point by photogrammetric
measurement
Akshay Jain

 Scale of a vertical photographs

 Tilt and relief displacement
 Stereoscopic vision
 Parallax in aerial stereoscope
 Effects of change of elevation and parallax
 Parallax bar
 Numerical

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

Objectives
1. Determining the scale of a vertical photograph and estimating
horizontal ground distances from measurements made on a
vertical photograph.
2. Using area measurements made on a vertical photograph to
determine the equivalent areas in a ground coordinate system.
Akshay Jain

3. Quantifying the effects of relief displacement on vertical aerial

photographs.
4. Determination of object heights from relief displacement
measurements.
5. Determination of object heights and terrain elevations by
measurement of image parallax.
6. Use of ground control points.
7. Mapping with aerial photographs.
8. Preparation of a flight plan to acquire aerial photography.

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

1
 2/11/2016

Akshay Jain

Parallax

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

A little test
 Stretch one arm and hold a pen in your hand. Close
one eye and look where the pen is located in its
environment. Close this eye and open the other.
 What do you observe?
Akshay Jain

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

Akshay Jain

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

2
 2/11/2016

This is what happened... the parallax

1. LEFT EYE 2. RIGHT EYE
Akshay Jain

The left image is seen by the left eye, the right image is seen by the right
eye. When you alternately close one eye, it looks as if the spoon is jumping
from one location to the other.
Question: On the background you see a bottle of water, which also changes
position. Compare its parallax with the parallax of the spoon. What's your
conclusion?

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

Answer
Akshay Jain

The parallax of the bottle is much smaller than the parallax of the spoon.

Conclusion: the parallax of closer objects is bigger than the parallax of distant
objects.

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

Parallax in remote sensing

 The same phenomenon appears when an airplane
(or a satellite) is flying over an area and taking
photographs from certain features (e.g. buildings)
with an interval of a few seconds (thus from a
Akshay Jain

different location).
 Have a look at the images below.

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

3
 2/11/2016

The IKONOS satellite took the following images of

these high buildings in Caïro (Egypt).
Akshay Jain

You get two images of the same building, but each one taken from a
different position. The apparent displacement of the building caused by the
change in the point of observation is the parallax.

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

Image Parallax
 The term parallax refers to the apparent change in
relative positions of stationary objects caused by a
change in viewing position.
 This phenomenon is observable when one looks a
Akshay Jain

objects through a side window of a moving vehicle.

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

Calculating the Height Out of a Parallax

 If two vertical aerial photographs are taken
consecutively along a flight line, with some overlap
of coverage between the two, the height can be
calculated.
Akshay Jain

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

4
 2/11/2016

In the following figure, a point A is imaged onto the two frame

photographs at al and ar, with the x coordinate origins at the principal
points (or perspective centre points, i.e. the sensors), O1 and O2.

The parallax at point A is expressed by:

Akshay Jain

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

 The parallax at point A can be shown graphically by

transferring the right image point to the left image as
shown in the next figure. We can use the parallax to
obtain information about the height of point A.
Akshay Jain

Observing the similarity between triangles (L1, L2, A)

and (L1, a1, a1) we may infer that:

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

Akshay Jain

Parallax at a point A. H is the distance from the sensor to a reference level.

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

5
 2/11/2016

 This confirms what we have already stated

intuitively, that parallax and proximity are inversely
related. Large parallax implies that the object is closer
to the sensor, i.e. the object has a great height, while
Akshay Jain

small parallax implies that the object is more distant,

i.e. the object has a lower height.

Parallax and proximity are inversely related. Large

parallax implies close proximity (meaning object's’
height is great) while small parallax implies a distant
object (meaning object’s height is low).

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

Parallax Bar
Akshay Jain

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

Object Height and Ground Coordinate

location from Parallax Measurement
 Figure shows overlapping vertical photographs of a
terrain point, A. Using parallax measurements, we
may determine the elevation at A and its ground
coordinate location.
Akshay Jain

 Referring to Figure (a), the horizontal distance

between exposure stations Land L' is called B, the air
base.
 The triangle in Figure (b) results from
superimposition of the triangles at L and L' in order
to graphically depict the nature of parallax Pa as
computed earlier algebraically. From similar
triangles La’xax (Figure b) and LAxL‘ (Figure a)

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

6
 2/11/2016

Akshay Jain

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

 Above equations are commonly known as the parallax

equations.
 In these equations, X and Y are ground coordinates of a
point with respect to an arbitrary coordinate system
Akshay Jain

whose origin is vertically below the left exposure station

and with positive X in the direction of flight; p is the
parallax of the point in question; and x and y are the
photocoordinates of the point on the left-hand photo.
 The major assumptions made in the derivation of these
equations are that the photos are truly vertical and that
they are taken from the same flying height. If these
assumptions are sufficiently met, a complete survey of
the ground region contained in the photo overlap area of
a stereopair can be made.

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

7
 2/11/2016

Numerical
 The length of line AB and the elevation of its
endpoints, A and B, are to be determined from a
stereopair containing images a and b. The camera
used to take the photographs has a 152.4-mm lens.
Akshay Jain

The flying height was 1200 m (average for the two

photos) and the air base was 600 m. The measured
photographic coordinates of points A and B in the
"flight line" coordinate system are xa=54.61 mm, xb=
98.67 mm, ya=50.80 mm, yb=-25.40 mm, x’a=-59.45
mm, and x’b =-27.39 mm. Find the length of line AB
and the elevations of A and B.

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

 In many applications, the difference in elevation

between two points is of more immediate interest than
is the actual value of the elevation of either point. In
such cases, the change in elevation between two
points can be found from
Akshay Jain

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

8
 2/11/2016

Stereoscopic Vision
Akshay Jain

Have you ever wondered how it is possible that man can see depth? And
why a rabbit can't?

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

Stereo Vision
 The location of both eyes is of great importance to
understand this marvelous phenomenon. While the
two eyes of a rabbit are at each side of the skull, our
eyes are located side-by-side on our face. This brings
Akshay Jain

on an overlap between the image seen by the right

eye and the one seen by the left eye. This overlap
makes stereo vision - the capability to perceive depth
- possible.

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

What is stereovision?
 Try to close one eye, and then the other. Each eye
perceives a view that differs slightly from the other.
This small difference is called the parallax. When
both images are transmitted to the back of the brain,
Akshay Jain

this parallax makes it possible to unite them into

one, three-dimensional image.
 We already learnt to calculate depths and heights
out of a parallax.

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

9
 2/11/2016

What is needed to make stereovision possible?

 The most evident instruments are of course our two
eyes that will put together two images in one three-
dimensional image. To see in stereovision, we also
need a pair of images that will be united to form a 3-
Akshay Jain

D image. An image is needed for the left eye and

another is to be seen by the right eye only. These two
images are called a stereopair. They have to be taken
by a sensor onboard an airplane or a satellite and to
include an overlap to make stereovision possible.

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

Stereoscope
Akshay Jain

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

Stereoscope
Akshay Jain

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

10
 2/11/2016

Two images of the same feature that are taken

from a different point of view form a stereopair.
Akshay Jain

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

Stereopairs
 Two images of the same area, taken from different
points of view are called a stereopair.
 In a stereopair, there is always an overlap (see red
squares), where stereovision is possible, because of
Akshay Jain

the present parallax. Outside these regions, you are

not able to see in 3D.

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

Viewing systems
 To get these two images merged together in a 3D
image, the observer has to see the pair of images
through a viewing system called a stereoscope.
 The stereoscope is a tool which guarantees that each
Akshay Jain

eye sees only one of the stereopairs. Each eye thus

sees a different image of the same scene, each image
taken from a slightly different point of view. When
the brain merges the two images together, the
observer experiences a 3D vision of the scene. This
can be realized manually or digitally.

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

11
 2/11/2016

Manual and Digital Stereoscope

Akshay Jain

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

How do we obtain stereopairs?

 When scientists want to map out a specific area, they
first create a flight map. Flight lines are laid out on
this flight map with a spacing that allows the
photographs to cover a common strip of ground.
Akshay Jain

This overlap between flight strips amounts to about

25% of the width of the area covered by the
photograph.
 Each photograph in the flight line covers an area
which overlaps the area covered by the previous
photograph by about 60%.

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

Akshay Jain

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

12
 2/11/2016

Satellites
 Satellites can make stereoscopic images in two ways:
1. A satellite takes one image sideways when it is
located at orbit (i) around the Earth. When it is one
orbit further (i+1) it makes another image of the same
Akshay Jain

area.

2. A satellite takes an image forward. A few seconds

later, it makes an image of the same area, backward.

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

Stereopair from Satellite

Akshay Jain

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

Orthophotos
 Do you think perspective projection is suitable for
constructing a topographic map?
 Take a look at images of Chicago and Washington
D.C. in the United States.
Akshay Jain

 Which image do you think will be the most useful

for creating a topographic map? Why?

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

13
 2/11/2016

Chicago, Illinois
Akshay Jain

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

Washington, D.C.
Akshay Jain

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

Chicago Case
 In the first situation, the plane/satellite takes an
image from a part of Chicago, forwards, backwards
or sideways from its current location (the
perspective centre (point P) in the air/in space. This
Akshay Jain

means that the image was made at an angle that is

less than 90°. As seen before, this projection is called
the perspective projection.

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

14
 2/11/2016

Perspective Projection
Akshay Jain

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

Washington D.C. Case

 The second image gives you the impression that this
is what you see when the plane/satellite (and thus
the perspective center) is located right above each
feature in the image, which means under an angle
Akshay Jain

that is exactly 90°.

Orthographic projection.

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

Orthorectification
 But in an aerial photograph or satellite image, which is always
a perspective projection, there is only one point in the image
where such a view is possible: the point just underneath the
perspective center/camera lens (point D in the first image). All
the other points in the image are taken under a certain angle,
Akshay Jain

which means that a distortion takes place.

 To be able to work with satellite images (direct and accurate
measurements of distances, angles, positions and areas), it is
much easier to have images like the second image of
Washington. Otherwise the image distortion has to be taken
into account in the calculations.
 The good news is that a process to correct the deformation
induced by the perspective projection,
called orthorectification, can be applied to remotely sensed
images. Orthorectified images are often called orthophotos.

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

15
 2/11/2016

Orthorectified Images
 In orthophotos and orthorectified images, each point
in the image gives you the impression that it was
taken at an angle of 90°. Therefore, there is no
distortion and each object is located on the map at
Akshay Jain

exactly the same relative position to each other as

they are on the surface of the Earth (except with a
change in scale).

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

Orthorectification
 Digital satellite images and aerial photographs play
an important role in the construction of maps or
visualization in a GIS. However, the images acquired
by Earth observation systems cannot be transferred
Akshay Jain

to maps or to GIS as such, because they are

geometrically distorted. The corrective process used
to remove distortion from a remotely sensed image
is called orthorectification. Without this process, you
would not be able to make direct and accurate
measurements of distances, angles, positions and
areas on remotely sensed images.

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

Orthorectification
Akshay Jain

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

16
 2/11/2016

What Causes the distortion?

 The topographical variations in the surface of the Earth
affect the relief displacement, i.e. the distance with
which features are displayed on the satellite or aerial
image. When the landscape is topographically very
Akshay Jain

diverse (very hilly or mountainous), you will find more

inherent distortion in the image.
 If the sensor is pointing directly at the nadir location, no
distortion will take place at nadir point. However, the
further away you get from nadir line in the image, the
bigger will be the image distortion. These terrain
distortions are of course amplified when the sensor takes
oblique images, i.e. when the image sensor is not
pointing directly at the nadir location.
CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

Distortion
 Other geometric distortions can affect the image, for
example, distortions due to errors in the satellite's
positioning on its orbit or to the fact that the Earth is
turning on its axis as the image is being recorded.
Akshay Jain

Again, these distortions are amplified even more

when the image is taken sideways.

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

Digital Elevation Model

Akshay Jain

3D DEM map of Pennsylvania

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

17
 2/11/2016

Digital Elevation Model

 A digital elevation model (DEM) is a digital
representation of a terrain surface. It is also known
as a digital terrain model (DTM).
 As the name implies, the elevation (also called
Akshay Jain

height or altitude) is the main parameter that is

displayed in the DEM. But before the elevation (z) at
a specific location can be determined, a reference
system for its horizontal position (x,y) has to be
pointed out. The most commonly used geographic
coordinate systems are UTM (Universal Transverse
Mercator) and the latitude-longitude coordinate
system.
CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

How do you construct a DEM?

 To create a DEM, the elevation (z) of every location in the
area is needed. Because it is not possible to survey them
all, only some locations, the so-called grid points, are
stipulated, where the elevation is accurately measured. It
means that at these locations, all three parameters
Akshay Jain

(x, y and z) are known.

 To determine the elevation of the remaining
coordinates, interpolation is needed. The closer together
the grid points are located, the more detailed the
information will be in the DEM. The details of the peaks
and valleys in the terrain will be better modelled with
small grid spacing than when the grid intervals are very
large. Elevations other than at the specific grid point
locations are not contained in the DEM. As a result peak
points and valley points which do not coincide with the
grid will not be recorded in the DEM.

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

Akshay Jain

Questions:
Have a look at the above figures. The red lines symbolize
DEMs trying to represent a terrain surface (in black).
1. Which of the DEMs fits the curves of the terrain more
accurately? Which interpolation is better and why?
2. What is indicated by the blue arrow?
CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

18
 2/11/2016

Answers
1. Because the grid points are closer, the bottom DEM
matches the real relief more precisely. In the top
figure, the grid points are too far away from each
other to get a good interpolation of the terrain.
Akshay Jain

2. The blue arrow indicates a peak point which was

not modelled at all by the DEM. This peak was
skipped by the DEM because it was not recorded
as a grid point prior to interpolation.

CE 211T Advanced Survey Unit III Photogrammetry PDPU-Gandhinagar

Thanks

19