MNG-MININGENGINEERING

SEM- V-SEMESTER
SUB CODE: 18MNG-505E[C]
SUB NAME: GEO-INFORMATICS AND ITS APPLICATIONS IN MINING
CHAPATER-4: PHOTOGRAMMETRY

Topic: 4.1 Explain the Historical development of Photogrammetry

Qn.1. Mention the development cycles of Photogrammetry.

Answer:
Development cycles of Photogrammetry:
(a).Plane table photogrammetry, from about 1850 to 1900
(b).Analog photogrammetry, from about 1900 to 1960
(c).Analytical photogrammetry, from about 1960 to present and
(d).Digital photogrammetry, which is just beginning to be a presence in the photogrammetric
industry.

Qn.2.Define the term Photogrammetry

Answer: Photogrammetry is the science and technology of obtaining reliable information about
physical objects and the environment through the process of recording, measuring and
interpreting photographic images and patterns of electromagnetic radiant imagery and other
phenomena.

Qn.3.Write an essay on historical development of Photogrammetry

Answer: HISTORY OF PHOTOGRAMMETRY:

This is the science of creating, producing image and photographs date back to the 1400’s believe
it or not, Leonard de Vince” is one of the earliest explorers of the science. Since then other
scientist continued this work on projective geometry mathematically. In 1525” Albercht duere”
used law of perspective drawing. From this same idea scientist began developing mathematical
principles of a perspective image using space reaction” (determining the coordinates of a point
from which a picture from two or more images) to find a point in space from which is made. The
actual relationship between projective geometry and photogrammetry was first developed by
Rostrums’ and Guido Haick in Germany in 1883.Since then the science went through four
development cycles.
ANALOG PHOTOGRAMMETRY (1910-1950):
The invention of the device ‘’Stereo comparator” by pullfrich, in 1909 marks the first revolution
of the photogrammetry, through which could make easy the work of users surprisingly, thanks to
the replacement of many mathematical calculations by opticalmechanical devices.
In 1911, the Austrians Theodore Scheimp flug creates a successful method of
rectification of aerial photographs. The analogue rectifiers are replaced by the famous analogical
stereo plotter, allowing stereoscopic vision using a stereoscopic pair. Many stereo plotters,
especially the swiss and germans, like wild, zeiss and kem had the opportunity to obtain
topographic charts with amazing precision. Here the information acquired is recorded on paper
or other media placed on top of a drawing table, which is connected mechanically or electrically
to stereo plotter. Obviously such devices, highly specific, require and deep training and the
‘’photogrammetist’’ becomes an essential part in photogrammetric process. We can consider yet
another classification-semi-analytical-in which the electronic data processing allowed the data
would be entered directly into systems Computer Aided Design (CAD), the drawing table was
eliminated, some other operations, in addition to registration, began to be assisted by computer.
The field work becomes easier by the introduction of the aerial triangulation process, allowing
deifying ground control points at the laboratory. Alongside these developments, starting appears
specific cameras, called metric cameras which have mechanisms, to print in the photo relevant
information about the image coordinate system, adding further the correctness of the made
measurements. A very good example is the fiducially marks. They are present in all photos and
their photos coordinates are determined in the laboratory; this is the time of producing and
empowering of a lot of organizations together all stakeholders in the area; the ISP ( International
Society For Photogrametry). Now the ISPRS (International Society for Photogrametry And
Remote Sensing) was founded in 1910 by E Dolezal; Austria.
ANALYTIC PHOTOGRAMMETRY (1951-1990): The invention/introduction of computer in
40;s began the transformation in the photogrammetric processes and the large calculations
needed , that the where replaced by mechanical devices, started to be executed on computer. The
first study in this area was developed in 1953 by Dr. Helmut Schmidt of the ballistics research
laboratory in Aberdeen, Maryland, USA. This study established the basis of analytical
photogrammetry, including the process matrix, the solutions by least squares (a type of statistical
method), the simultaneous solution using multiple images and complete analysis of error
propagation.
In 1957, the finish Uki Helava develops the concept of analytical stereo plotter, using
servomechanisms to measure the coordinates of fiducial marks on images. Perform all other
calculations simplifying very much the process. The first analytical stereo plotter was presented
in the congress of the ISP (now ISPR) in 1976. Thereafter, such equipment revotionalized the
concept of photogrammetry, allowing serial triangulation in larger books and the use of ordinary
cameras (non-metric)
Digital photogrammetry (1990-date):
The digital photogrammetry was born in the 80s, having as a great innovation the use of digital
images as a primary data source. The digital image can be acquired directly from a digital
camera, or even by scanning or analog. In the 1990a, this branch of photogrammetry has been
used extensively, thanks to the development of computers with sufficient capacity for the
interactive processing of digital images, generating high volume of data.

Qn.4. Write a short note on Digital photogrammetry.

Answer: Digital photogrammetry (1990-date): The digital photogrammetry was born in the
80s, having as a great innovation the use of digital images as a primary data source. The digital
image can be acquired directly from a digital camera, or even by scanning or analog. In the
1990a, this branch of photogrammetry has been used extensively, thanks to the development of
computers with sufficient capacity for the interactive processing of digital images, generating
high volume of data.

Qn.5. Write a short note on Analytic photogrammetry

Answer: ANALYTIC PHOTOGRAMMETRY (1951-1990): The invention/introduction of
computer in 40;s began the transformation in the photogrammetric processes and the large
calculations needed , that the where replaced by mechanical devices, started to be executed on
computer. The first study in this area was developed in 1953 by Dr. Helmut Schmidt of the
ballistics research laboratory in Aberdeen, Maryland, USA. This study established the basis of
analytical photogrammetry, including the process matrix, the solutions by least squares (a type of
statistical method), the simultaneous solution using multiple images and complete analysis of
error propagation.
In 1957, the finish Uki Helava develops the concept of analytical stereo plotter, using
servomechanisms to measure the coordinates of fiducial marks on images. Perform all other
calculations simplifying very much the process. The first analytical stereo plotter was presented
in the congress of the ISP (now ISPR) in 1976. Thereafter, such equipment revotionalized the
concept of photogrammetry, allowing serial triangulation in larger books and the use of ordinary
cameras (non-metric).

Topic: 4.2 Define Aerial photography

Qn.6.Mention any four parts of an Aerial camera.

Answer: Parts of an Aerial camera:
An aerial camera consists of three basic parts: a body, a cone, and a magazine.
The body has following constituent parts: Filter, Front lens, Shutter, Diaphragm and Rear lens.
The magazine has three constituents: Supply reel/ spool, Take up reel/spool and Platen.

Qn.7. Define the term Aerial photography

Answer: Aerial photography (or airborne imagery) is the taking of photographs from an aircraft
or other flying object. Platforms for aerial photography include fixed-wing aircraft, helicopters,
unmanned aerial vehicles (UAVs or "drones"), balloons, blimps and dirigibles, rockets, pigeons,
kites, parachutes, stand-alone telescoping and vehicle-mounted poles. Mounted cameras may be
triggered remotely or automatically.

Qn.8. Explain the Aerial photography

Answer: Aerial photography is - as it sounds - the process of taking photographs from the air,
but there is more to it than simply using a light aircraft or helicopter and flying up to take
photographs. There are many elements to an aerial survey that must be considered to ensure that
the data is useful enough to extrapolate whatever is being investigated. It is often difficult to see
elements of the landscape on the ground, features can easily be missed, and what might seem like
an insignificant bump from ground level can become more significant in a wider context; some
landscape types are difficult to access on foot so aerial photographs are vital to study and map
them. They have been used as a method of landscape studies for over a century, especially in
archaeology and researchers have learnt much about the world around us; its applications today
are broad and coupled with the growing technology of GIS (geographic information systems),the
potential means that the method will not become obsolete any time soon. Aerial photographs are
taken in two basic forms and both have different uses and applications: oblique and vertical.
Even today in an age of high quality digital imaging, black and white images are preferred –
partly because they are cheaper but also partly because the contrast of black, white and greys
makes it easier to pick out features.
Oblique
These images are usually taken at an angle, typically 45 degrees but as they are often taken
manually, they can be whatever angle gives the best view of the feature or landscape. The
oblique image is primarily used in archaeology to take a wider context of a feature and the area
around it, and also to give depth. Nearly always taken at a much lower elevation than the vertical
image and in few numbers, its application is fairly limited and often taken for a specific purpose.
There is a problem in perspective because the farther away a feature is, the smaller it will appear:
nearer objects of comparable size appear larger than those that are farther away so it is often best
to take a selection or to use a frame of reference on the ground for perspective purposes. These
images are taken from small fixed-wing aircraft and helicopters and are perfectly suited for
monitoring erosion of features and monuments throughout the year and over the course of many
decades.
Qn.9. Write an essay on the historical development of Aerial photography?
Answer: Aerial photography was first practiced by the French photographer and balloonist
Gaspard-Félix Tournachon, known as "Nadar", in 1858 over Paris, France. However, the
photographs he produced no longer exist and therefore the earliest surviving aerial photograph is
titled 'Boston, as the Eagle and the Wild Goose See It.' Taken by James Wallace Black and
Samuel Archer King on October 13, 1860, it depicts Boston from a height of 630m. Aerial view
by Cecil Shadbolt, showing Stonebridge Road, Stamford Hill, and Seven Sisters Curve, part of
the Tottenham and Hampstead Junction Railway, taken from 2,000 feet (610 m) on 29 May 1882
- the earliest extant aerial photograph taken in the British Isles. Kite aerial photography was
pioneered by British meteorologist E.D. Archibald in 1882. He used an explosive charge on a
timer to take photographs from the air. The same year, Cecil Shadbolt devised a method of
taking photographs from the basket of a gas balloon, including shots looking vertically
downwards. One of his images, taken from 2,000 feet (610 m) over Stamford Hill, is the earliest
extant aerial photograph taken in the British Isles. A print of the same image, An Instantaneous
Map Photograph taken from the Car of a Balloon, 2,000 feet high, was shown at the 1882
Photographic Society exhibition. Frenchman Arthur Batut began using kites for photography in
1888, and wrote a book on his methods in 1890. Samuel Franklin Cody developed his advanced
'Man-lifter War Kite' and succeeded in interesting the British War Office with its capabilities.
The first use of a motion picture camera mounted to a heavier-than-air aircraft took place on
April 24, 1909.

Topic: 4.3 What is Vertical photography

Qn.10.Define the term Vertical photography

Answer: Vertical photographs are taken straight down. They are mainly used in photogrammetry
and image interpretation. Pictures that will be used in photogrammetry are traditionally taken
with special large format cameras with calibrated and documented geometric properties.
Qn.11. Explain the Vertical photography
Answer: Vertical photographs are the most common type of aerial photograph for remote
sensing and aerial survey purposes. They can be scaled, allowing objects and distances to be
measured, aiding in their identification. When viewed in stereo, vertical photographs can give
information about the height or the vertical characteristics of landmarks and buildings.
Vertical aerial photographs can provide very useful information, in conjunction with maps and
other sources, when searching for unexploded ordnance or assessing property boundaries, for
example.

Topic: 4.4. Define Oblique Aerial photography

Qn.12.Define the term Oblique images

Answer: Oblique images: These images are usually taken at an angle, typically 45 degrees but
as they are often taken manually, they can be whatever angle gives the best view of the feature or
landscape. The oblique image is primarily used in archaeology to take a wider context of a
feature and the area around it, and also to give depth.

Qn.13. Explain Oblique Aerial photography

Answer: Oblique Aerial Photograph

is Photographs taken at an angle. If they are taken from a low angle earth surface–aircraft, they
are called low oblique and photographs taken from a high angle are called high or steep oblique.
The technical definition of vertical aerial photograph is photography looking down on a subject
at an angle of less than or equal to three degrees from the vertical. In other words vertical aerial
photograph is defined as an aerial photograph technique where the shots are taken from directly
above the subject of the image.

The following are advantages of oblique aerial photograph over vertical aerial photograph:
1.Oblique aerial photograph may use inexpensive cameras because they are not used for
photogrammetric or precision purpose. Determination of feature elevations is more accurate
using oblique aerial photograph than vertical aerial photograph.
2. An oblique aerial photograph covers more ground area than vertical aerial photograph taken
from the same altitude and with the same focal length.
3. If an area is frequently covered by cloud layer, it may be too low and/or impossible to take
aerial vertical photograph but there may be enough clearance for oblique coverage.
4. Oblique aerial photograph have more natural view because we are accustomed to seeing the
ground features obliquely. For example tall objects such as bridges, building towers and others
will be more recognizable because the silhouettes of these objects are visible.
5. Objects that are under tree or under other tall objects may not be visible on vertical aerial
photograph if they are viewed from above. Also some objects such as ridges, cliffs, caves and
others of the same reflection may not show on the vertical aerial photograph if they are directly
beneath the camera.

Topic: 4.5 Classify the Aerial film and Digital cameras

Qn.14. Define the term Panchromatic film

Answer: Panchromatic film: Panchromatic, more often termed black and white, is the most
commonly encountered film employed for photogrammetry. The sensitive layer consists of silver
salt (bromide, chloride, and halide) crystals suspended in a pure gelatin coating which sits atop a
plastic base sheet. Visible light waves react with the silver particles in the emulsion, causing a
chemical reaction that creates a gray-scale image. The emulsion is sensitive to the visible (0.4- to
0.7-μm) portion of the electromagnetic spectrum that is detected by the human eye.

Qn.15. Briefly explain the Aerial film cameras

Answer: Aerial film is similar in construction to the film popularly used in handheld 35-mm
cameras. It comes in rolls that are 10 in. wide and range in length from 200 to 500 ft.
Types of Film
Although there are a number of aerial films in use, many serve unique situations. Two commonly
utilized films employed in plan metric and/or topographic digital mapping are panchromatic and
natural color. These two films plus infrared and false color form the basic media used in image
analysis procedures
Panchromatic
Panchromatic, more often termed black and white, is the most commonly encountered film
employed for photogrammetry. The sensitive layer consists of silver salt (bromide, chloride, and
halide) crystals suspended in a pure gelatin coating which sits atop a plastic base sheet. Visible
light waves react with the silver particles in the emulsion, causing a chemical reaction that
creates a gray-scale image. The emulsion is sensitive to the visible (0.4- to 0.7-μm) portion of the
electromagnetic spectrum that is detected by the human eye.
Colour
Natural color film is also called true color or color. The multilayer emulsion is sensitive to the
portion of the electromagnetic spectrum that is visible to the human eye. There are three layers of
gelatin containing sensitized dyes, one each for blue (0.4–0.5 μm), green (0.5–0.6 μm), and red
(0.6–0.7 μm) light. Green and red layers are also sensitive to blue wavelengths. Visible light
waves first pass through and react with the blue layer and then pass through a filter layer which
halts further passage of the blue rays. Green and red waves pass through this barrier and sensitize
their respective dyes, causing a chemical reaction and thus completing the exposure and creating
a true color image.
Infrared
Current aerial infrared film is offered as two types: black and white infrared and color infrared.
Qn.16. Explain the Aerial film and Digital cameras
Answer: Aerial film is similar in construction to the film popularly used in handheld 35-mm
cameras. It comes in rolls that are 10 in. wide and range in length from 200 to 500 ft.
Types of Film
Although there are a number of aerial films in use, many serve unique situations. Two commonly
utilized films employed in plan metric and/or topographic digital mapping are panchromatic and
natural color. These two films plus infrared and false color form the basic media used in image
analysis procedures
Panchromatic
Panchromatic, more often termed black and white, is the most commonly encountered film
employed for photogrammetry. The sensitive layer consists of silver salt (bromide, chloride, and
halide) crystals suspended in a pure gelatin coating which sits atop a plastic base sheet. Visible
light waves react with the silver particles in the emulsion, causing a chemical reaction that
creates a gray-scale image. The emulsion is sensitive to the visible (0.4- to 0.7-μm) portion of the
electromagnetic spectrum that is detected by the human eye.
Colour
Natural color film is also called true color or color. The multilayer emulsion is sensitive to the
portion of the electromagnetic spectrum that is visible to the human eye. There are three layers of
gelatin containing sensitized dyes, one each for blue (0.4–0.5 μm), green (0.5–0.6 μm), and red
(0.6–0.7 μm) light. Green and red layers are also sensitive to blue wavelengths. Visible light
waves first pass through and react with the blue layer and then pass through a filter layer which
halts further passage of the blue rays. Green and red waves pass through this barrier and sensitize
their respective dyes, causing a chemical reaction and thus completing the exposure and creating
a true color image.
Infrared
Current aerial infrared film is offered as two types: black and white infrared and color infrared.
Black and White Infrared
Infrared film is also known as black and white infrared. The emulsion is sensitive to green (0.54–
0.6 μm), red (0.6–0.7 μm), and part of the near infrared (0.7–1.0 μm) portions of the spectrum
and renders a gray-scale image. Positive images appear quite like panchromatic film, except that
water and vigorous vegetation tend to register as darker gray to black. The film structure
resembles panchromatic with the exception that the emulsion sensitivity range is shifted upward,
eliminating blue wavelengths and including a portion of the near infrared. In the past this film
was used extensively in vegetation and water studies, but its popularity seems to be declining in
favor of color infrared.
Color Infrared
Color Infrared film is commonly termed false color. The multilayer emulsion is sensitive to
green (0.5–0.6 μm), red (0.6–0.7 μm), and part of the near infrared (0.7–1.0 μm) portions of the
spectrum. A false color image contains red/pink hues in vegetative areas, with the color
depending upon the degree to which the photosynthetic process is active. It also images water in
light blue/green to dark blue/black hues, depending on the amount of particulates suspended in
the water body. Clean water readily absorbs near infrared radiation. As the amount of foreign
particulates increases, the near infrared rays reflect increasingly more of these particles. The film
structure resembles natural color, except that the blue sensitive layer is eliminated and replaced
by a layer that reacts to a portion of near infrared (0.7–1.0 μm).

Topic: 4.6 What are the components of Aerial cameras

Qn.17. Briefly explain the important specifications of Aerial camera

Answer: IMPORTANT SPECIFICATIONS:
Almost every aerial camera has to fulfill certain specification:
1. The focal length of the camera should be normally 6, 8.25, or 12 inches.
2.The camera should expose 9 ´ 9-inch image negatives; the focal plane and fiducial marks be
permanently fixed on the photographs.
3.The platen against which the film is pressed at the time of exposure should not depart by more
than ± 0.0005 inches.
The radial distortion should normally not exceed ± 0.003mm out to 30O and ±0.10 mm at 35O.
Qn.18. Write a short note on Patrs of an Aerial camera
Answer: PARTS OF AN AERIAL CAMERA:
An aerial camera consists of three basic parts: a body, a cone, and a magazine.
The body has following constituent parts:
1. Filter
2. Front lens
3. Shutter
4. Diaphragm
5. Rear lens
The magazine has three constituents:
1. Supply reel/ spool
2. Take up reel/spool
3. Platen
The cone is intermediate part of the camera which keeps the camera body and magazine at a
definite distance apart from each other.

Topic: 4.7 Define Camera calibration.

Qn.19. What is Camera calibration

Answer: After manufacture and prior to use, aerial cameras are carefully calibrated to determine
precise values for a camera constants. These constants, referred to as the elements of interior
orientation, are needed so that accurate data can be determined from photographs. Camera
calibration methods may be classified into one of three basic categories: 1. Laboratory methods,
2. Field methods, and 3. stellar methods.
Camera calibration of interior orientation information:
• 1. Equivalent focal length
• 2. Calibrated focal length
 • 3. Average radial lens distortion
• 4. X,Y coordinates
• 5. Distances between opposite fiducial marks
• 6.Angle of intersection of fiducial lines
• 7. Flatness of focal plane

Topic: 4.8 What is Mosaicing

Qn.20. List out the main processes of document mosaicing

Answer: The document mosaicing can be divided into four main processes.
-Tracking
-Feature detecting
-Correspondences establishing
-Images mosaicing

Qn.21.Write a short note on Feature detection process of mosaicing

Answer: Feature detecting for efficient matching
Feature detection is the process of finding the transformation that aligns one image with another.
There are two main approaches for feature detection.
Feature-based approach: Motion parameters are estimated from point correspondences. This
approach is suitable for the case that there is plenty supply of stable and detectable features.
Featureless approach: When the motion between the two images is small, the motion
parameters are estimated using optical flow. On the other hand, when the motion between the
two images is large, the motion parameters are estimated using generalized cross-correlation.
However, this approach requires a computationally expensive resource. Each image is segmented
into a hierarchy of columns, lines, and words to match the organized sets of features across
images. Skew angle estimation and columns, lines and words finding are the examples of feature
detection operations.

Qn.22.Write a short note on document tracking mosaicing process

Answer: Tracking (simple correlation process):
In this process, the motion of the document slide under the camera is coarsely tracked by the
system. Tracking is performed by a process called simple correlation process. In the first frame
of snapshots, a small patch is extracted from the center of the image as a correlation template as
shown in Figure 1. The correlation process is performed in the four times size of the patch area
of the next frame. The motion of the paper is indicated by the peak in the correlation function.
The peak in the correlation function indicates the motion of the paper. The template is re
sampled from this frame and the tracking continues until the template reaches the edge of the
document. After the template reaches the edge of the document, another snapshot is taken and
the tracking process performs repeatedly until the whole document is imaged. The snapshots are
stored in an ordered list to facilitate pairing the overlapped images in later processes.

Qn.23. Write an essay on Components of Aerial cameras

Answer: PARTS OF AN AERIAL CAMERA:
An aerial camera consists of three basic parts: a body, a cone, and a magazine.
The body has following constituent parts:
1. Filter
2. Front lens
3. Shutter
4. Diaphragm
5. Rear lens
The magazine has three constituents:
1. Supply reel/ spool
2. Take up reel/spool
3. Platen
The cone is intermediate part of the camera which keeps the camera body and magazine at a
definite distance apart from each other.

FUNCTIONING:
When the camera is focused at infinity, the lens gathers light rays reflected from the objects and
transmits them in an orderly fashion to the light-sensitive areas known as the film. The shutter
serves to regulate the amount and duration of light reaching the film when making an exposure
and it has a millimeter opening which overcomes the jerks and blurring caused by the moving
aircraft and to capture the whole picture. In addition to a fast and effective shutter, the camera
should have a high-grade lens which should admit sufficient light at the required shutter speed
and must be adaptable to the particular camera design and the film used and should have low
distortion characteristics.
TYPES OF AERIAL CAMERA:
There are various types of aerial cameras but the most commonly used are single lens camera.
Aerial cameras can be classified in a number of ways:
1. By lens type: single lens, multiple lenses.
2. By angular field: normal angle up to 750; wide angle 750 to 1000; and, super wide angle
1000 and above.
3. By focal length: short up to 6 inches; normal 6 to 12 inches; and, long more than 12 inches.
4. By use: reconnaissance, mapping, and special
The selection of any camera would depend upon the need and use. For instance, the cameras
used for low altitude high-speed photography must have wide angle lenses, fast shutter, image
motion compensation magazines and short cycle times. On the other hand, for higher altitude
photography long-focal length cameras are used to provide a reasonable scale and resolution.
Commonly used cameras are Zeiss EMKA 15/23 and Wild RC-8. These work on both
panchromatic and colored films. Other include Metric camera (having no focusing), Stereo
metric camera (with focusing), and Armature camera (used for less accuracy).

IMPORTANT SPECIFICATIONS:
Almost every aerial camera has to fulfill certain specification:
1. The focal length of the camera should be normally 6, 8.25, or 12 inches.
2. The camera should expose 9 ´ 9-inch image negatives; the focal plane and fiducial marks be
permanently fixed on the photographs.
3. The platen against which the film is pressed at the time of exposure should not depart by more
than ± 0.0005 inches.
The radial distortion should normally not exceed ± 0.003mm out to 30O and ±0.10 mm at 35O.

Qn.24. Write an essay on Mosaicing

Answer: Document mosaicing is a process that stitches multiple, overlapping snapshot images
of a document together to produce one large, high resolution composite. The document is slid
under a stationary, over-the-desk camera by hand until all parts of the document are snapshotted
by the camera’s field of view. As the document slid under the camera, all motion of the
document is coarsely tracked by the vision system. The document is periodically snapshotted
such that the successive snapshots are overlap by about 50%. The system then finds the
overlapped pairs and stitches them together repeatedly until all pairs are stitched together as one
piece of document.
The document mosaicing can be divided into four main processes.
-Tracking
-Feature detecting
-Correspondences establishing
-Images mosaicing
Tracking (simple correlation process)
In this process, the motion of the document slid under the camera is coarsely tracked by the
system. Tracking is performed by a process called simple correlation process. In the first frame
of snapshots, a small patch is extracted from the center of the image as a correlation template as
shown in Figure 1. The correlation process is performed in the four times size of the patch area
of the next frame. The motion of the paper is indicated by the peak in the correlation function.
The peak in the correlation function indicates the motion of the paper. The template is resampled
from this frame and the tracking continues until the template reaches the edge of the document.
After the template reaches the edge of the document, another snapshot is taken and the tracking
process performs repeatedly until the whole document is imaged. The snapshots are stored in an
ordered list to facilitate pairing the overlapped images in later processes.
Feature detecting for efficient matching Feature detection is the process of finding the
transformation that aligns one image with another. There are two main approaches for feature
detection.
Feature-based approach : Motion parameters are estimated from point correspondences. This
approach is suitable for the case that there is plenty supply of stable and detectable features.
Featureless approach : When the motion between the two images is small, the motion
parameters are estimated using optical flow. On the other hand, when the motion between the
two images is large, the motion parameters are estimated using generalized cross-correlation.
However, this approach requires a computationally expensive resources. Each image is
segmented into a hierarchy of columns, lines, and words to match the organized sets of features
across images. Skew angle estimation and columns, lines and words finding are the examples of
feature detection operations.
Correspondences establishing
The two images are now organized in hierarchy of linked lists in following structure :
image=list of columns
row=list of words
column=list of row
word=length (in pixels)
At the bottom of the structure, the length of each word is recorded for establishing
correspondence between two images to reduce to search only the corresponding structures for the
groups of words with the matching lengths.
Images mosaicing
Given the list of corresponding points of the two images, finding the transformation of the
overlapping portion of the images is the next process. Assuming a pinhole camera model, the
transformation between pixels (u, v) of image 1 and pixels (u0, v0) of image 2 is demonstrated
by a plane-to-plane projectivity.