ArticlePDF Available

Determining the Geographical Location of Image Scenes based on Object Shadow Lengths

Authors:

Abstract and Figures

Many studies have addressed various applications of geo-spatial image tagging such as image retrieval, image organisation and browsing. Geo-spatial image tagging can be done manually or automatically with GPS enabled cameras that allow the current position of the photographer to be incorporated into the meta-data of an image. However, current GPS-equipment needs certain time to lock onto navigation satellites and these are therefore not suitable for spontaneous photography. Moreover, GPS units are still costly, energy hungry and not common in most digital cameras on sale. This study explores the potential of, and limitations associated with, extracting geo-spatial information from the image contents. The elevation of the sun is estimated indirectly from the contents of image collections by measuring the relative length of objects and their shadows in image scenes. The observed sun elevation and the creation time of the image is input into a celestial model to estimate the approximate geographical location of the photographer. The strategy is demonstrated on a set of manually measured photographs.
Content may be subject to copyright.
1
Determining the Geographical Location of
Image Scenes based on Object Shadow
Lengths
Frode Eika Sandnes
Faculty of Engineering, Oslo University College, P.O. Box 4, St. Olavs plass,
0130 Oslo, Norway
frodes@hio.no
This is a revised and extended version of a paper presented at The Pacific Rim
Conference on Multimedia, PCM2009, in Bangkok, December, 2009.
Abstract Many studies have addressed various applications of geo-spatial image tagging such as
image retrieval, image organisation and browsing. Geo-spatial image tagging can be done
manually or automatically with GPS enabled cameras that allow the current position of the
photographer to be incorporated into the meta-data of an image. However, current GPS-equipment
needs certain time to lock onto navigation satellites and these are therefore not suitable for
spontaneous photography. Moreover, GPS units are still costly, energy hungry and not common in
most digital cameras on sale. This study explores the potential of, and limitations associated with,
extracting geo-spatial information from the image contents. The elevation of the sun is estimated
indirectly from the contents of image collections by measuring the relative length of objects and
their shadows in image scenes. The observed sun elevation and the creation time of the image is
input into a celestial model to estimate the approximate geographical location of the photographer.
The strategy is demonstrated on a set of manually measured photographs.
Keywords: geo-spatial tagging, image content analysis, image classification.
1 Introduction
Automatic image classification, labelling and retrieval are active research topics
[29, 30]. Most photographers do not have the time and patience to manually
catalogue single photographs and label these with textual descriptions. Instead,
most users are often able to memorize approximately when a photo was taken, say
“during the summer of 2008”, or “in the winter holiday after the September 11
event”. Moreover, users will have few problems associating a particular image
with a location, such as “our holiday in Puerto Rico”, “the business trip to Cape
Town” or “the PCM 2009 conference in Bangkok”. These are all possible because
cameras not only store the images recorded by the camera chips but also store the
time and date when the photos were taken using a digital clock built into the
camera. Some cameras also store camera settings such as exposure time, aperture,
2
focal distance, focal length, etc, using EXIF (Exchangeable Image File Format)
[1] initiated by the Japan Electronics and Information Technology Industries
Association (JEITA). This meta-information can also be used to organize images
[2].
Geo-spatial information is an emerging image attribute that is used in addition to
the time and date of an image. Combined time and geo-spatial attributes make it
easier to organise, retrieve and browse large image collections [3, 4]. Moreover,
image collections are growing rapidly and often viewed on mobile devices.
Falling costs have resulted in most people owning digital cameras, and the quality
of the camera equipment is constantly improving. Currently, even mobile phone
cameras have megapixel resolution. Low cost digital storage has eliminated cost
and time barriers previously associated with the development of film.
Still, GPS technology is not commonplace in most digital cameras as they add to
the cost in a very competitive market. Moreover, although the idea of using GPS
technology is attractive in theory, it may not always be practical. A photographer
may have to react spontaneously to a given situation and quickly take a shot.
However, GPS enabled devices often need certain amount of time to lock onto the
available overhead GPS satellites. In fact, the process of obtaining a reasonable
GPS reading can sometimes take several minutes. Next, imagine that very
response GPS enabled cameras became commonplace, then there will still be huge
collections of digital photographs in existence taken with older digital cameras
without geo-spatial capabilities. Finally, the current GPS-infrastructure is reaching
the end of its lifetime and one does not have any guarantees for publically
available satellite navigation systems in the future [5].
1.1 Direct sun elevation measurements
GPS technology is a relatively new phenomenon. Prior to GPS technology
navigation and positioning was achieved using the position of celestial bodies
such as the sun, moon and the stars. During days with clear skies the sun provides
a good reference point for estimating ones position. Based on the time of year, the
sun follows a sinusoidal path across the skies relative to an observer on earth. On
the northern hemisphere the sun goes up in the east and sets in the west and is
located at a southern direction at midday. On the southern hemisphere the sun
goes from east to west via a northern route. Generally, the elevation of the sun is
3
higher at midday for small latitudes compared to high latitudes where the
maximum elevation of the sun is lower. Moreover, during winter the elevation is
lower than the summer, and while it is winter on the northern hemisphere it is
summer on the northern hemisphere and vice versa.
Seafarers have exploited this phenomenon for hundreds of years. For instance, the
sextant was used to measure the elevation of the sun above sea level by aligning
two adjustable views. One view was centred on the horizon and another view was
centred on the sun, such that the two views were aligned. Then, an accurate
angular reading of the suns elevation was taken. Next, the height of the observer
above sea level was compensated for. By the means of an accurate watch, a
compass and an astronomical almanac the position of the observer was estimated
with a very high accuracy of close to 0.1 nautical miles which is approximately
200 meters.
These traditional celestial navigation techniques have inspired researchers
working on autonomous robot navigation where a digital camera was used to
measure the approximate elevation the sun as a kind of digital sextant [6]. Related
research includes the development a sun sensor [22].
A lens is usually characterised in terms of its focal length f. A simplified
explanation of focal length is how much magnification a lens provides. A lens
with a large focal length magnifies an image more than a lens with a smaller focal
length. However, with more magnification the lens field of view is smaller. The
field of view covered by a lens with focal length f is given by
f
d
a2
tan2 1
(1)
where d represents the width of the image sensor inside the camera. Classic 35
mm film has a dimension of 36 x 24 mm, while digital camera sensors often are
smaller. For instance, cameras in the Nikon’s DX series have dimensions of about
23.6 x 15.5 mm, Cannon APS-C has dimensions of 22.2 x 14.8 mm, and pocket
camera sensors can be as small as 2.4 x 1.8 mm (1/6” sensors). Usually the lenses
are rectilinear, that is, all straight edges in the scene appear straight in the captured
image. The field of view can be measured along the horizontal (width), vertical
(height) or along the diagonal. It is the dimensions of the sensor (or digital) film
that determines the field of view along the vertical and horizontal dimensions. A
35 mm camera with a 50 mm lens will therefore have a horizontal view of 46.8
4
degrees and a vertical view of 27 degrees. It has been shown that the lens focal
length for a camera can be determined using a sequence of outdoor images where
the position of the sun is hand labelled [7].
Given a camera configuration with a resolution of Px x Py pixels and a field of
view of Vx x Vy degrees along the horizontal and vertical positions, respectively.
Then the degrees per pixel are given by:
y
y
x
xP
V
P
V
a
(2)
The vertical degrees per pixel should be approximately the same along the
horizontal and vertical axis. Given an optimal image scene comprising clear skies,
a sun and a distinct horizon, the distance in pixels between the sun and the horizon
are easily measured, and hence the elevation e of the sun can be calculated as
horizonsun yyae
(3)
Where ysun is the vertical pixel value for the centre of the sun and yhorizon a
representative vertical pixel value of the horizon assuming the camera is level.
Several methods for horizon extraction have been proposed, including the use of
orientation projection [8, 9]. These are robust methods aimed at micro aircraft
control with unfocused rapidly moving images. Given the elevation of the sun and
the current solar time an astronomer’s almanac can be used to determine the
geographical location [13].
The direct sun elevation measurement technique is not well suited for the analysis
of digital image collections. First, the calculations are dependent on the
characteristics of the physical camera design. Second, most camera lenses have a
limited field of view and will only work when the sun is at low elevations. For
example, with a 50 mm lens and 35 mm digital film the maximum theoretical
elevation is 26 degrees. With a 100 mm lens and 35 mm digital film the maximum
theoretical elevation is 14 degrees, and for a 200 mm lens and 35 mm digital film
the maximum theoretical elevation is 6 degrees. Next, with the exception of
beautiful sunrises and sunsets, it is uncommon to take direct photographs of the
sun. Finally, although accurate horizon detection algorithms exist for small
aircrafts flying at certain altitudes, it is much harder to determine the altitude from
a photographers perspective as he or she may be located in a city, in a valley or
next to other tall objects that obstructs the view of the horizon [20].
5
1.2 Indirect sun elevation measurements
Direct sun observations can be avoided by measuring the sun elevation indirectly.
In particular, the position of the sun has also been measured indirectly by
investigating the lighting condition of a scene [25], represented using the exposure
level. The lighting conditions are related to the elevation of the sun, where in
general solar noon is the brightest time of day. The exposure level can be
computed using the aperture, shutter speed and film speed settings that many
digital cameras store in the image EXIF headers [1, 2]. Experiments have shown
that a brightness representation of the suns trajectory can be sufficiently mapped
for image collections. Based on these trajectories rough estimates of solar noon
and day-lengths can be made. Solar noon and day-length measurements can again
be used to estimate the longitude and altitude of the observer. This approach has
been demonstrated to yield a longitudinal accuracy of 15 degrees and a latitudinal
accuracy of 30 degrees with arbitrary holiday photo collections [25]. A problem
with this strategy is that it requires a sufficiently large set of outdoor images with
a sufficiently large temporal spread. For images without exposure metadata, it has
been demonstrated that a very rough indication of longitude can be determined by
simply taking the mean time for a sequence of images within a 24 hour window as
the solar noon. The achieved accuracy for arbitrary collections of holiday photos
was about 30 degrees [26]. An advantage of both these indirect methods is that
they also work under cloudy conditions, and the latter strategy even works
indoors.
1.3 Webcam measurements
Another branch of related research attempts to determine the geographical
location of webcams [23, 26, 28]. Webcams are often used to acquire sequences
of regularly spaced images for monitoring purposes. The cameras are usually
located in a fixed location and often pointing in a constant direction. On the
downside, few webcams store meta-information in EXIF headers and analysis can
therefore only be performed using actual image contents. Webcam image
sequences have been used to determine the relative position of webcams and their
orientation [23, 24]. Moreover, an accuracy of about 2 degrees was achieved using
a contents-based intensity measure of webcam images sampled every 5 to 11
minutes [28]. This approach allowed the sunrise and sunsets to be determined, and
6
hence the solar noon and length of day could be calculated. However, webcam
images represent a special case and webcam techniques are not applicable to
general image collections.
1.4 Landmark recognition
Another novel approach to geo-tagging involves automatically recognizing known
landmarks in image scenes. Given knowledge about the location of the landmarks
the location of the image scene can therefore be inferred [21]. Such strategies
clearly depend on both an extensive landmark database and a powerful landmark
matching algorithm.
1.5 Object-shadow lengths and sun elevation
This study proposes a new strategy for deriving the geographical origin of image
scenes based on both the image contents and image meta-information. The
proposed strategy relies on the fact that the lengths of shadows cast by vertical
objects on horizontal surfaces indirectly reveal the elevation of the sun. If such
sun elevation measurements are obtained together with the time at which
photographs were taken it is possible to derive the geographical location where
the images were captured. There are several locations at which one can observe
the sun at a given elevation at a given time. Therefore, up to three images taken at
different times at the same location are used to identify a single and unique
geographical location. This study investigates the practicality, reliability and
accuracy of such object-shadow length sun elevation measurements for
determining geographical location of image scenes. Although this strategy will
not work on cloudy days it has potential for much greater accuracy than previous
indirect methods based on scene brightness.
H
L
e
sun
object
shadow
surface
Fig 1. The relationship between the sun elevation e, object height H and shadow length L.
7
2 Shadows and sun elevation
Shadows provide an indirect clue to the elevation of the sun as the sun at a high
elevation will cast a short shadow while the sun at a low elevation will cast a long
shadow. Given an object with a height H and a shadow with length L, the
elevation e of the sun is simply
L
H
e1
tan
(8)
This is illustrated in Fig. 1. A convenient property of this equation is that it is
based on a ratio and any units associated with the object and shadow length
measurements are cancelled. Hence, the shadow based sun elevation
measurements are close to independent of the technical properties of the camera
and the relative dimensions of the scene with the exception of distortions caused
by low quality lenses.
Fig 2. The sun position is the point on earth where the sun elevation is 90 degrees, namely the
position with latitude δ (sun declination angle) and longitude w (solar angle). For an observer the
elevation angle e depends on the observers position, i.e., latitude φ and longitude λ.
Next, it can be shown that the relationship between the elevation of the sun e and
the geographical location of the observer (see Fig. 2) is given by:
we coscoscossinsin)sin(
(9)
where φ is the latitude of the observer, w is the sun angle of the observer and δ is
the declination of the sun at the given date which can be approximated by:
)10(
365
2
cos -0.4092797 M
(10)
Here, the declination of the sun is represented in radians and M denotes the day of
the year. The constant 0.4092797 represents the maximum declination angle of
8
the sun, or earth tilt, in radians (23.45 degrees) that occurs during the two solstices
(see Fig. 3). Note that this is a rough approximation of the sun declination angle,
i.e., a simple sinusoidal with a period of 365 days, and that more accurate
approximations exist. However, the author’s experimentation has shown that this
expression provides sufficient accuracy for the purpose of this study.
axis
equator
23.5 °
axis
equator
23.5 °
axis
equator
23.5 °
axis
equator
23.5 °
Sun
Summer solstice
Approx. June 21
Winter solstice
Approx. Dec. 21
Spring equinox
Approx. Mar. 20
Autumn equinox
Approx. Sept. 23
Fig 3. The northern hemisphere is more exposed to the sun during the summer and the southern
hemisphere is exposed to the sun during the winter and the maxima occur during the two solstices
as the earth’s tilt is then parallel to the direction of the sun. Both hemispheres are equally exposed
to the sun during the two equinoxes as the earth’s tilt is then perpendicular to the direction of the
sun.
Next, the longitude λ of the observer is related to the solar time tsun as follows
12
utcsun tt
(11)
and solar time tsun is related to the sun angle w as follows:
)12(
12
180 sun
tw
(12)
Given an elevation measurement e1 at UTC time t1 one can find all observation
points with the given sun elevation for the given time. In this study we traversed
the Earths surface with a resolution of 1 degree, giving, 360 × 180 points and
stored all locations in L1 which satisfied the sun elevation criteria for the given
time. For high elevations the possible locations form a circle-like shape on the
Earth’s surface as shown in Fig. 4.
9
-90
-70
-50
-30
-10
10
30
50
70
90
-180 -160 -140 -120 -100 -80 -60 -40 -20 020 40 60 80 100 120 140 160 180
7.59 utc
10.48 utc
13.18 utc
Fig. 4. The location traces for three elevation observations from Cape Town, South Africa at 7.59,
10.48 and 13.18 UTC during February 27, 2009. The three traces cross approximately in one
location.
In order to get a more accurate fix on the actual location a second sun elevation e2
at a different image taken at time t2 is obtained, giving rise to a second trace of
locations L2 (see Fig. 4). These, two traces cross in two locations (φ1, λ1) and (φ2,
λ12) one on the southern and one on the northern hemisphere.
In order to determine which of the two estimated locations that represents the true
location a third sun elevation e3 from a third image taken at time t3 is needed. This
gives rise to a third trace of location points L3. Then, in most situations there will
be only be one point where all the three traces L1, L2 and L3 cross simultaneously,
namely the true location (φ, λ) of the observer. Note that also the correct
hemisphere is determined in these cases.
-90
-60
-30
0
30
60
90
-180 -150 -120 -90 -60 -30 030 60 90 120 150 180
7 utc
9 utc
11 utc
13 utc
15 utc
Fig 5. The location traces for sun elevation observations at Oslo, Norway during the March 20
equinox at 7:00, 9:00, 11:00, 13:00 and 15:00 UTC, respectively. Note that the different traces
cross in two points one on the southern and northern hemisphere.
10
-90
-60
-30
0
30
60
90
-180 -150 -120 -90 -60 -30 030 60 90 120 150 180
9 utc
11 utc
13 utc
Fig. 6. The location traces for sun elevation observations at Oslo, Norway during January 1 at
9:00, 11:00 and 13:00 UTC, respectively. The traces only cross in one point the location of the
observations.
The feasibility of this approach is dependent on the season. It will work especially
well during the winter and during the summer where the declination of the sun is
large, while it will work less well during the spring and autumn when the
declination of the sun is small. With a large declination the length of day is very
different on the two hemispheres and the sun elevation paths are very distinct (see
Fig. 6). On the contrary, with a small sun declination the differences between the
sun elevation paths on the hemisphere are small and it is harder to distinguish
between the two (see Fig. 5). In other words, the approach works best closest to
the two solstices (generally 21st of June and 21st of December) and the strategy
will not be able to distinguish between the two hemispheres during the two
equinoxes (approximately 20th March and 23rd September). This hemisphere
ambiguity is illustrated in Fig. 5. With small sun declinations it is necessary with
additional clues in order to determine which hemisphere the observer was located
at.
The ability to successfully identify the correct hemisphere is also dependent on
the angle between the latitude and the declination of the sun. With a large solar
angle and a latitude close to the declination of the sun angle, it is more difficult to
determine on which hemisphere the observer is located, while this is much easier
when the angle between the latitude and the sun declination is large. Yet, if the
observers latitude is close to the declination of the sun and an observation is
made close to the solar noon, that is, with a small solar angle, then the location
can also be determined quite accurately as the sun can only be observed at
elevations of close to 90 degrees at a limited area on the Earth’s surface.
11
Moreover, traces for sun elevations taken at different times will also only cross in
one point. This is illustrated in Fig. 7. The plot shows that all the traces only cross
through one point. Therefore, images taken at latitudes close to the sun declination
line can be determined with one image if the sun angle is small and with two
images otherwise. The plot shows that the diameter of the trace 12:30 is only 15
degrees, while at 12:00 the trace is simply one point. One hour before and after
noon the diameter of the traces are 30 degrees and grows with 30 degrees for each
hour in either direction away from the solar noon.
-90
-60
-30
0
30
60
90
-180 -150 -120 -90 -60 -30 030 60 90 120 150 180
7:00 utc
8:00 utc
9:00 utc
10:00 utc
11:00 utc
12:30 utc
Fig. 7. Sun elevation traces for observations at (0, 0) during an equinox.
2.1 Land test
Previous sections have demonstrated that it may be difficult to determine the
correct hemisphere when images are taken close to the equinoxes or if shadows
from only two images are used. For this purpose a simple land test is proposed. It
comprises mapping the two points onto a simple world map to determine if the
points hit land or water. The one that hits land is chosen.
Imagine for example that two images are taken in Oslo, Norway (59.9 degrees
north, 10.7 degrees east) during the spring equinox of March 20th. These will yield
the coordinates (59.9°, 10.7°) and (-59.9°, 10.7°). Fig. 8 shows these coordinates
plotted onto a world map. Clearly, the former is located at Oslo, while the latter is
located in the ocean south of the African continent. Unless the photograph was
taken onboard a ship it is natural to reject the latter coordinate and conclude that
the coordinate on the northern hemisphere is correct. By inspecting the world map
in Fig. 8 it is obvious that the simple map test works for most locations in
Northern Europe, North America and Asia. This is because approximately 70% of
the Earth’s surface is covered in water.
12
-90
-60
-30
0
30
60
90
-160 -130 -100 -70 -40 -10 20 50 80 110 140 170
Fig. 8. Using a map to resolve a hemisphere ambiguity. The coordinate for the southern
hemisphere is rejected as it does not refer to a land area. The map is taken from Wikipedia
(Creative Commons).
Fig. 9 summarizes the proposed strategy for determining the geographical location
of a set of image scenes. Input to the algorithm are three sun elevation
measurements obtained from the object-shadow length ratios, the times the three
images were captured and the date of the event. The output of the algorithm is the
approximate geographical location of the place the images where captured.
Coordinate findLocation(Angle elevationa, Time timea,
Angle elevationc, Time timeb,
Angle elevationc, Time timec,
Date date)
begin
tracea locationsWithSunElevation(elevationa, timea, date)
traceb locationsWithSunElevation(elevationb, timeb, date)
tracec locationsWithSunElevation(elevationc, timec, date)
location tracea traceb tracec
If date=equinox then
return landTest(location)
else
return location
end
Set locationsWithSunElevation(Angle observedElevation, Time time, Date date)
begin
declination sunDeclination(date) // Eq. 10
for longitude -180 to 180 step resolution
for latitude -90 to 90 step resolution
begin
elevation ← calcElevation(latitude, longitude, declination, time) // Eq. 9 and 12
if observedElevation ≈ elevation then
trace.add(Coordinate(latitude, longitude))
end
return trace
end
Fig. 9. Algorithm for determining the location of image scenes based on object-shadow lengths.
13
2.2 Automatic object-shadow length measurements
This study focuses on how to determine the approximate geographical location
given a set of object-shadow length measurements. Obtaining accurate object-
shadow length measurements is indeed a non-trivial problem as one has to
identify objects, identify shadows and determine which objects relate to which
shadows. Therefore, only a rough speculation on how this may be achieved is
attempted here. Inspiration is drawn from the literature which contains several
accounts of work related to shadow detection [14, 15]. For instance shadow
detection has been successfully applied to video based on colour models [16].
Segmentation of objects and background in outdoor images has also been studied
[17] as well as shadows in aerial photographs [18, 19].
An image collection may be large and advanced processing of all the images is
unrealistically time-consuming. A natural first step is therefore to identify suitable
image candidates, that is, images that are likely to have shadows. This is simply
achieved by using the exposure attributes stored in EXIF-headers, including the
aperture f (f-number), shutter speed s and film speed iso. Based on these the
exposure level EV can be determined [31, 32]:
100
loglog 2
2
2iso
s
f
EV
(13)
Then outdoor images taken on a sunny day with sufficient shadows should have
an exposure value of approximately 12 or more. If EXIF information is not
available a content based strategy can be used to identify suitable candidate
images although that will be computationally more demanding than simply
inspecting the EXIF-information. Several content-based strategies for classifying
outdoor and indoor images have been proposed in the literature, for instance using
colour space histograms [10] and support vector machines [11]. Moreover,
attempts at extracting information from daytime images of the skies [12] have
been proposed.
Next, candidate images can be separated into their hue and brightness
components. Objects may be identified and segmented in the hue plane [27], and
shadows identified and segmented in the brightness plane. Having obtained these
segments the object lengths and shadow lengths can be measured.
14
This procedure can be repeated for several images and statistical approaches can
be used to assess what shadow measurements that should be accepted and which
ones that should be rejected.
Clearly, the outlined strategy is challenging as one may easily detect false objects
and false shadows and thus end up with erroneous sun elevation measurement.
Therefore, further research is needed to identify robust extraction strategies.
2.3 Time and date assumptions
The strategy presented herein assumes that all images are consistently time-
stamped with date and time. Further, it is assumed that the time-zone is known
such that the times can be converted to UTC (Coordinated Universal Time). All
the calculations presented herein are represented in UTC. Most owners set their
camera to the time zone of their home country. Few users bother to change the
time of their cameras when travelling to a different country in a different time
zone. Since the camera clocks usually have their own battery one may assume that
for most users the time will be set to the same time-zone for the entire lifetime of
the camera and that potential time drifts will affect all images equally.
2.4 Image scene assumptions
The shadow model is also based on two further assumptions. First, the viewing
plane is approximately level. If standing in a slope such as on the side of a hill the
shadow angle calculations would require the model to take the slope into
consideration. Given a slope of s degrees and a shadow of length L cast up the
slope, then the error in the shadow due to the slope is E = L - L cos(s).
Second, the model assumes that all the objects are completely vertical with
straight lines. Curved or tilting objects will cast more complex shadows and an
angle extraction algorithm will have to take information about the scene into
consideration. When a curved and tilted object is combined with a sloping surface
the extraction of shadow information is even more complex. One strategy would
be to classify images according to how tilted the ground is and the tilted or curved
the objects are. Images with such characteristics can then be eliminated from the
shadow extraction procedure as their geometry is too complex for simple analysis
procedures.
15
Fig 10. The manual 3-point sun-elevation measurement procedure.
3 Experimental evaluation
3.1 Test suite
To assess the technique proposed herein a series of photographs taken at two
campuses of Cape Peninsula University of Technology in Cape Town, South
Africa during February 27, 2009 were used. This was a sunny day with clear skies
and hence distinct shadows. The collection was photographed by the author, but
without this experiment in mind. The sample therefore represents an arbitrary and
natural image collection. A Sony DSC-F828 digital camera with 8 megapixel
resolution and a zoom lens was used. First the image collection was manually
inspected and a set of 8 photographs were selected. The following criteria had to
be satisfied: The image scene had to contain a visible object and this object had to
cast a visible shadow. The objects had to be vertical and straight. Only images
where the shadows perceivably fell approximately perpendicular to the camera
direction were selected to minimize image projection distortions. That is, images
with shadows going straight left or right were selected. For each of the selected
images Microsoft Paint was used to measure the exact pixel locations of three
16
object-shadow feature points, namely the top of the object, the point connecting
the object and the shadow and the shadow end point. These three points make up
an L-shape, or inverted L shape as illustrated in Fig. 10. In this example the
rubbish bin makes up the object and the shadow is cast on the right side of the bin.
Next, EXIF-information, including the time and date of the photograph and the
focal length used, were extracted using Microsoft Office Picture Manager. The
images used and the associated feature points are illustrated in Fig. 11. Table 2
lists test suite details including the UTC time, measured elevation, the length of
the measured shadow vector and the focal length of the lens used (degree of wide
angle or zoom).
17
Fig. 11. The images used in the experimental evaluation. Detailed attributes of these images are
given in Table 2. The image resolution is 3264 × 2448 pixels.
18
Table 1. Image test suite used in the experiments.
UTC time
UTC
decimal
time
(hours)
measured
elevation
(degrees)
shadow
vector
length
(pixels)
focal
length
(mm)
07:33:10
7.6
34.4
2056.4
7.1
07:35:40
7.6
34.1
332.6
7.1
09:02:00
9.0
56.6
1225.4
7.1
10:21:10
10.4
62.8
478.8
7.1
10:29:13
10.5
63.0
995.4
7.1
10:36:31
10.6
73.6
526.1
28.1
13:10:58
13.2
52.4
411.8
7.1
13:11:13
13.2
60.0
300.8
36.5
The coordinate 33.9 degrees south, 18.8 degrees east was used to represent Cape
Town in this experiment. The date of the image collection is the 58th day of the
year when the declination of the sun is approximately -9.1 degrees. Hence, there is
a significant difference between the hemispheres. This date is 21 days away from
the spring equinox with no hemisphere difference and 68 days away from the
winter solstice when there are maximum seasonal differences between the
hemispheres.
Table 2. Accuracy of latitude and longitude estimates.
2 images
3 images
accuracy
rank
images
latitude
longitude
latitude
longitude
high-high
6, 7, 8
3, 4, 5
-2.1°
-1.2°
-2.1°
-1.2°
high-medium
5, 6, 7
7, 5, 4
-0.1°
5.8°
-0.1°
5.8°
medium-medium
4, 5, 6
1, 7, 5
0.9°
6.8°
0.9°
6.8°
medium-low
3, 4, 5
2, 1, 7
0.9°
6.8°
0.9°
6.8°
low-low
1, 2, 3
8, 6, 2
16.9°
12.8°
16.9°
12.8°
3.2 Geographical accuracy
Table 2 summarizes the result obtained with the proposed strategy. These results
both demonstrate the accuracy of the strategy and the effects of varying the
accuracy of the elevation measurements that are the input to the algorithm. First,
the images were ranked according to the accuracy of their measured elevation
accuracy. Then, a sliding window of 3 images was run through the ranking list to
generate 5 sets of images with varying accuracy. The table therefore lists a
linguistic description of accuracy, the rank of the images used, the actual index of
19
the images used and the latitude and longitudes obtained with both the two and
three image techniques.
The results show that the overall best estimate had a latitudinal error of 2.1
degrees and longitudinal error of 1.2 degrees. Then, as the accuracy of the sun
elevation measurements decreased the largest error for this dataset was 16.9
degrees latitude and 12.8 degrees longitude. These results are superior to those
obtained using image intensity [25] and matches the accuracy obtained using
webcam image sequences [28].
Note that both the 2-image and 3-image strategies yield the same accuracy. The
only effective difference between the two techniques is that the 3-image method
was capable of automatically resolving the correct hemisphere and the 2-image
solutions had to be resolved manually.
These results are much less accurate than the accuracy offered by GPS receivers.
However, the purpose of this strategy is not to navigate, or survey landmass. The
purpose is to geo-tag images and an accuracy of approximately 2 degrees suffices
for uniquely distinguishing continent and even country. It would, however, be
interesting to investigate if the accuracy could be further improved by using
images taken with this strategy in mind, that is, images where the photographer
ensures that a clear shadow and its object is captured such that they occupy a
majority of the image view and that the shadow is perpendicular to the camera.
0
10
20
30
40
50
60
70
80
7 9 11 13
Sun elevation (degrees)
UTC Time (hours)
observed
theoretic
Fig 12. The measured and theoretical sun elevations for the eight images used in this experiment.
3.3 Shadow measurement accuracy
Fig. 12 shows that the observed sun elevations follow the theoretical sun
elevations with a few exceptions. The first two elevation measurements are too
low and the 6th and last elevation measurements are too large.
20
There are several sources of error in the above experiment. First, the camera clock
may not be completely accurate. However, an inspection of the camera revealed
that the clock was accurate to 2 minutes from the actual time. Still, the time will
only affect the longitude. If the time is off by one hour the longitudinal error will
be 15 degrees, for every minute of clock error the longitude error is 0.25 degrees
and every second of time inaccuracy affects the longitude by 0.004 degrees.
Therefore, an error of up to 2 minutes could have affected the longitude by up to
half a degree. Note that an unsynchronized clock will not affect the latitude
estimates since all the images are correctly spaced in relative time.
Fig 13. Perspective distortion affects the perceived shadow angle.
Distortions caused by camera projections may be a source of error (see Fig. 13).
Although, all the shadows are perceived to be perpendicular to the camera
direction it may not be the case in practice. In particular, for images taken with the
zoon, that is, shadows that are further away will visually appear more
perpendicular than shadows that are taken with wider lens configuration and that
are closer to the camera. This is particularly noticeable if the plane of the shadow
is close in height to the observer. Fig. 13 illustrates how the shadows on a plane
below the observer appear less perpendicular than shadows on a plane on similar
height to the observer. The effect is that these shadows are erroneously observed
as too short. This effect is further amplified by camera object distance. This
hypothesis is backed up by the results where sun elevation errors appear to
correlate with the level of zoom (focal length). The two measurements with the
largest error, that is, the sixth image and the eight image are both taken with
zoom, namely focal lengths of 28.1 mm and 36.5 mm, respectively, where the
latter yields the largest sun elevation error. The other images are taken using a
21
wide angle lens with a focal length of 7.1 mm. By inspecting the last image,
showing a student walking down a set of stairs, one sees that the measured
shadow falls on a plateau. The projection makes the shadow appear perpendicular
to the camera direction and the width of the plateau appears narrow. But, an
inspection of the image as a whole will reveal that this plateau in fact is quite wide
and that the shadow is at a slight angle. If one was standing closer one may have
observed that the direction of this shadow is far from perpendicular to the camera
angle. Consequently, the shadow measurement is too short compared to the object
height resulting in a sun elevation measurement that is too high. This error is
confirmed by the results in Fig. 12 where the measured sun elevation is 11.4
degrees higher than the theoretical sun elevation. The measured shadow length
was 154 pixels while the actual length should have been 235 pixels. The
measurement was therefore short by about 81 pixels, or 34%. Future work should
therefore introduce some measure to compensate for projection distortions. This
involves identifying potential inaccurate shadow measurements by taking the
distance into consideration where the distance is related to the focal length of the
lens, the actual length of the shadow in number of pixels and the position of the
shadow within a scene. A small shadow may indicate a shadow further away. A
shadow closer to the middle of a scene (low-medium y-value), that is, closer to the
horizon, is likely to be further away from the camera compared to a shadow
towards the bottom of a scene (high y-value) that is likely to be closer to the
camera.
0
10
20
30
40
50
60
70
7 9 11 13
Sun elevation(degrees)
UTC time (hours)
Simple model
Elaborat e model
Fig. 14. A comparison of the simple and elaborate sun elevation models.
3.4 Celestial model accuracy
The celestial model used in this study is simplistic as it is purely based on the
geometric properties of the sun and earth orbits. Advantages of this model include
22
that it is simple to implement, easy to describe and involves little computational
effort. However, other more elaborate and complex models exist that take other
factors into consideration such as atmospheric refraction [33]. Fig. 14 illustrates
differences between the simple and a more elaborate model. The data for the
elaborate model was acquired using an online sun-elevation calculator
(http://www.satellite-calculations.com/Satellite/suncalc.htm) that is implemented
according to a procedure described in [33]. The plot seems to suggest a minor
time discrepancy, that is, the simple model is slightly ahead in time of the more
elaborate model.
When comparing the simple and elaborate model with the actual measurements it
was found that the simple model yielded a mean sun elevation error of 4.9 degrees
(SD=3.6) and the elaborate model resulted in a mean sun elevation error of 3.9
degrees (SD=3.9). Hence, the elaborate model had an overall better fit to the
measurements compared to the simple model, although the spread in error was
also larger. Therefore, for any real applications of this approach the simple model
should be replaced with a more elaborate celestial model such as the one
described in [33]. Note that the strategy presented herein is general and works
with any celestial model.
4 Conclusions
A framework for determining the location a series of photographs based on the
contents of the images was presented. The elevation of the sun is determined
indirectly using the shadows cast by vertical objects. The advantage of shadow
based sun elevation extraction is that it can be performed without knowledge
about the optical properties of the camera or the absolute scale of objects in the
scene. Experimental results revealed that the location of images could be found
with an accuracy of down to 2 degrees in latitude and longitude given shadow
measurements with an error below 2 degrees of sun elevation. The meter-level
accuracy provided by GPS technology is usually not needed for image browsing
and cataloguing applications as an overall positioning accuracy of a few degrees is
sufficient to identify approximately where in the world the photographs are taken.
The strategy therefore has potential for content based geo-spatial information
retrieval. However, its success is reliant on the progress of future research into
automatic accurate object-shadow length measurement algorithms.
23
References
1. P. Alvarez, Using Extended File Information (EXIF) File headers in Digital Evidence
Analysis, International Journal of Digital Evidence 2(3) (2004).
2. C.-J. Jang, J.-Y. Lee, J.-W. Lee, and H.-G. Cho, Smart Management System for Digital
Photographs using Temporal and Spatial Features with EXIF metadata, presented at 2nd
International Conference on Digital Information Management, pp. 110-115, 2007.
3. S. Ahern, M. Naaman, R. Nair, and J. Hui-I Yang, World explorer: visualizing aggregate
data from unstructured text in geo-referenced collections, presented at 7th ACM/IEEE-CS
joint conference on Digital libraries, pp. 1-10, 2007.
4. D. Carboni, S. Sanna, and P. Zanarini, GeoPix: image retrieval on the geo web, from
camera click to mouse click, presented at Proceedings of the 8th conference on Human-
computer interaction with mobile devices and services, pp. 169-172, 2006.
5. GAO, GLOBAL POSITIONING SYSTEM: Significant Challenges in Sustaining and
Upgrading Widely Used Capabilities, United States Government Accountability Office
2009.
6. F. Cozman and E. Krotkov, Robot localization using a computer vision sextant, presented
at IEEE International Conference on Robotics and Automation, 1995.
7. J. F. Lalonde, Narasimhan, S.G., Efros, A.A.:, Camera parameters estimation from hand-
labelled sun positions in image sequences., Robotics Institute, Carnegie Mellon
University. Technical Report CMU-RI-TR-08-32, 2008.
8. G.-Q. Bao, S.-S. Xiong, and Z.-Y. Zhou, Vision-based horizon extraction for micro air
vehicle flight control, IEEE Transactions on Instrumentation and Measurement 54(3), pp.
1067-1072, 2005.
9. S. M. Ettinger, C. Nechyba, and P. G. lfju, Towards Flights autonomy: Vision-based
horizon detection for micro air vehicles, presented at IEEE International Conference on
Robotics and Automation, 2002.
10. M. Szummer and R. W. Picard, Indoor-outdoor image classification, presented at IEEE
International Workshop on Content-Based Access of Image and Video Database, 1998.
11. N. Serrano, A. Savakis, and A. Luo, A computationally efficient approach to
indoor/outdoor scene classification, presented at 16th International Conference on Pattern
Recognition, 2002.
12. J.-F. N. Lalonde, S. G. Efros, A. A., What does the sky tell us about the camera?,
presented at European Conference on Computer Vision, 2008.
13. J. J. Michalsky, Astronomer's Almanac algorithm (19502050), Solar Energy 40(3), pp.
227-235, 1988.
14. T. Horprasert, D. Harwood, and L. S. Davis, A Statistical Approach for Real-time Robust
Background Subtraction and Shadow Detection, presented at IEEE ICCV, 1999.
15. M. D. Levine and J. Bhattacharyya, Removing shadows, Pattern Recognition Letters
26(3), 251-265, 2005.
24
16. P. KaewTraKulPong and R. Bowden, An Improved Adaptive Background Mixture Model
for Realtime Tracking with Shadow Detection, presented at 2nd European Workshop on
Advanced Video Based Surveillance Systems, AVBS01, 2001.
17. S. Lefèvre, L. Mercier, V. Tiberghien, and N. Vincent, Multiresolution Color Image
Segmentation Applied to Background Extraction in Outdoor Images, presented at IS&T
European Conference on Color in Graphics, Image and Vision, 2002.
18. Y. Li, T. Sasagawa, and P. Gong, A system of the shadow detection and shadow removal
for high resolution city aerial photo, presented at XXth ISPRS Congress, 2004.
19. J. M. Wang, Y. C. Chung, C. L. Chang, and S. W. Chen, Shadow detection and removal
for traffic images, presented at IEEE International Conference on Networking, Sensing
and Control, 2004.
20. F. E. Sandnes, "Sorting holiday photos without a GPS: What can we expect from
contents-based geo-spatial image tagging?," Lecture Notes on Computer Science, vol.
5879, no., pp. 256-267, 2009.
21. Y.-T. Zheng, Z. Ming, S. Yang, H. Adam, U. Buddemeier, A. Bissacco, F. Brucher, T.-S.
Chua, and H. Neven, "Tour the world: Building a web-scale landmark recognition
engine," in the proceedings of IEEE Conference on Computer Vision and Pattern
Recognition (CVPR 2009), pp. 1085-1092, 2009.
22. A. Trebi-Ollennu, T. Huntsberger, Y. Cheng, and E. T. Baumgartner, "Design and
analysis of a sun sensor for planetary rover absolute heading detection.," IEEE
Transactions on Robotics and Automation, vol. 17, no. 6, pp. 939-947, 2001.
23. N. Jacobs, S. Satkin, N. Roman, R. Speyer, and R. Pless, "Geolocating Static Cameras,"
in the proceedings of IEEE 11th International Conference on Computer Vision (ICCV
2007), pp. 1-6, 2007.
24. N. Jacobs, N. Roman, and R. Pless, "Toward Fully Automatic Geo-Location and Geo-
Orientation of Static Outdoor Cameras," in the proceedings of IEEE Workshop on
Applications of Computer Vision, pp. 1-6, 2008.
25. F. E. Sandnes, “Where was that photo taken? Deriving geographical information from
image collections based on temporal exposure attributes”, Multimedia Systems, 2010
(accepted).
26. F. E. Sandnes, “Unsupervised and Fast Continent Classification of Digital Image
Collections using Time”, in Proceedings of ICSSE 2010, IEEE CS P ress, 2010 (to
appear).
27. Yo-Ping Huang, Tsun-Wei Chang, Yen-Ren Chen and Frode Eika Sandnes, "A Back
Propagation based Real-Time License Plate Recognition System", International Journal of
Pattern Recognition and Artificial Intelligence, Vol. 22, No. 2, pp. 233-251, 2008.
28. F. E. Sandnes, “A Simple Content-based Strategy for Estimating the Geographical
Location of a Webcam”, in Proceedings of PCM2010, Lecture Notes on Computer
Science, 2010 (to appear).
29. Wei Huang, Yan Gao and Kap Luk Chan, A Review of Region-Based Image Retrieval,
Journal of Signal Processing Systems, DOI:10.1007/s11265-008-0294-3.
25
30. Daniel Heesch and Maria Petrou, Markov Random Fields with Asymmetric Interactions
for Modelling Spatial Context in Structured Scene Labelling, Journal of Signal Processing
Systems, DOI: 10.1007/s11265-009-0349-0.
31. L. A. Jones and H. R. Condit, "The Brightness Scale of Exterior Scenes and the
Computation of Correct Photographic Exposure," Journal of the Optical Society of
America, vol. 31, no. 11, pp. 651-678, 1941.
32. S. F. Ray, "Camera Exposure Determination," in The Manual of Photography:
Photographic and Digital Imaging, R. E. Jacobson, S. F. Ray, G. G. Atteridge, and N. R.
Axford, Eds.: Focal Press, 2000.
33. P. Schlyter, Computing planetary positions - a tutorial with worked examples,
Downloaded March 26, 2010 from http://www.stjarnhimlen.se/comp/tutorial.html#5
Frode Eika Sandnes received a B.Sc. in computer science from the University of Newcastle upon
Tyne, U.K., and a Ph.D. in computer science from the University of Reading, U.K. He is currently
a Professor in the department of Computer Science at Oslo University College, Norway. His
research interests include multimedia processing, error-correction and human computer
interaction.
... Many researchers proposed and described these equations and presented architectural examples to estimate shadow behavior. In 2011, Sandnes Frode Eika determined the geographical location of image scenes based on object shadow lengths and knowing the time of the captured image [7]. In 2013, W. Seo, C. Yoon, J. S. Jang, S. ...
... The results from the first equation model (7) are tested in Fig. 4 with real-time to check the reliability of the introduced model. Hence, Figure 4 shows the first model in the red curve as a shadow angle function with the recorded shadow angle represented by black mark points. ...
Article
In this study, models are presented to estimate the time from an object shadow angle and estimate the shadow length from daytime. A stick length of 10 cm used to measure the shadow length and shadow angle from 7 am to 5 pm for 1st, 11th, 12th, 13th, 21st, 22nd, and 30th September 2019. A new algorithm presented to extract the shadow length and angle automatically of work the images quickly and with high accuracy. The presented algorithm results for both shadow length and shadow angle show a good matching of the measured data. The introduced models' results, for daytime and shadow length, have good agreement with the real data. The highest value of Normalize mean square error (NMSE), used as a criterion for daytime data (from the first model) and real daytime, was 0.029%. The (NMSE) for the real shadow length and shadow length (online) was 0.011% while the value (NMSE) between real shadow length with the resulting model was less than 0.010%. This small percentage shows a good matching in the estimation of the daytime and shadow length of an outdoor object in the different day-times. The presented models can generalize globally.
... Su yüzeylerinin filtrelenmesi sonucunda elde edilen görüntülere en yüksek benzerlik sınıflandırması (MLC) ve destek vektör makinaları (SVM) yöntemleri uygulanarak bitki gölgeleri sınıflandırılmıştır (Qiao ve ark., 2017). Sandnes (2011), güneşin pozisyonu ve gölge uzunluğunu kullanarak, sensörün yaklaşık konumunu tahmin etmiştir. Huang ve Chen (2009) hava platformlarından çekilen videolarda gölgelerin tespit edilebilmesi için yeni bir algoritma geliştirilmiştir. ...
Article
Full-text available
Yüksek çözünürlüklü multispektral görüntüler tarım alanlarının izlenmesinde oldukça yararlı bilgiler sunmaktadır. Fakat bu görüntülerdeki gölge alanları spektral yansıma oranı ve termal verileri direk olarak etkilemektedir. Bu nedenle gölge alanlarının tespit edilmesi ve filtrelenmesi, yüksek çözünürlüklü görüntülere dayalı olarak gerçekleştirilen çalışmalardaki başarı oranını arttıracaktır. Görüntülerde bulunan gölge alanları sınıflandırma yöntemleri ile tespit edilmektedir. Fakat, bu yöntemlerin insansız hava aracından (İHA) elde edilen yüksek çözünürlüklü multispektral görüntülerde kullanımı üzerine gerçekleştirilmiş araştırma sayısı oldukça azdır. Bu çalışmanın amacı üç farklı görüntü sınıflandırma yönteminin (eğitimli sınıflandırma, multispektral görüntü ile sınıflandırma ve sınıf olasılığı) İHA’ dan elde edilen multispektral görüntülerdeki gölge alanlarını tespit etmedeki başarısının değerlendirilmesidir. Her bir sınıflandırma yönteminin başarısı, görüntüde manuel yöntem ile belirlenen gölge alanları ile değerlendirilmiştir. Çalışma sonucunda bitki gölgeleri en hassas (%94) kırmızı kenar multispektral görüntüsü kullanılarak gerçekleştirilen sınıflandırma ile elde edilirken, en düşük hassasiyet (%74) ise eğitimli sınıflandırma yöntemi ile hesaplanmıştır.
... Physical-based methods Sandnes (2011) used the sun position and shadow length to approximately estimate the geolocation of the sensor. Huang and Chen (2009a) presented a physical approach for detecting the shadows in video imagery and showed that the proposed method can effectively identify the shadows in three challenging video sequences. ...
Article
Full-text available
Significant efforts have been made recently in the application of high-resolution remote sensing imagery (i.e., sub-meter) captured by unmanned aerial vehicles (UAVs) for precision agricultural applications for high-value crops such as wine grapes. However, at such high resolution, shadows will appear in the optical imagery effectively reducing the reflectance and emission signal received by imaging sensors. To date, research that evaluates procedures to identify the occurrence of shadows in imagery produced by UAVs is limited. In this study, the performance of four different shadow detection methods used in satellite imagery was evaluated for high-resolution UAV imagery collected over a California vineyard during the Grape Remote sensing Atmospheric Profile and Evapotranspiration eXperiment (GRAPEX) field campaigns. The performance of the shadow detection methods was compared and impacts of shadowed areas on the normalized difference vegetation index (NDVI) and estimated evapotranspiration (ET) using the Two-Source Energy Balance (TSEB) model are presented. The results indicated that two of the shadow detection methods, the supervised classification and index-based methods, had better performance than two other methods. Furthermore, assessment of shadowed pixels in the vine canopy led to significant differences in the calculated NDVI and ET in areas affected by shadows in the high-resolution imagery. Shadows are shown to have the greatest impact on modeled soil heat flux, while net radiation and sensible heat flux are less affected. Shadows also have an impact on the modeled Bowen ratio (ratio of sensible to latent heat) which can be used as an indicator of vine stress level.
... Studies have demonstrated that shadow trajectories inferred from multiple outdoor images can be used to determine the geolocation of stationary cameras [19], [24]. Sandnes [35] uses the relative lengths of objects and their shadows in images to estimate the sun altitude angle and further estimates the geolocation of the camera. Jacobs et al. [36] analyze the correlations of camera images with satellite images and other images with known locations to determine where these camera images were taken. ...
Preprint
The contextual information (i.e., the time and location) in which a photo is taken can be easily tampered with or falsely claimed by forgers to achieve malicious purposes, e.g., creating fear among the general public. A rich body of work has focused on detecting photo tampering and manipulation by verifying the integrity of image content. Instead, we aim to detect photo misuse by verifying the capture time and location of photos. This paper is motivated by the law of nature that sun position varies with the time and location, which can be used to determine whether the claimed contextual information corresponds with the sun position that the image content actually indicates. Prior approaches to inferring sun position from images mainly rely on vanishing points associated with at least two shadows, while we propose novel algorithms which utilize only one shadow in the image to infer the sun position. Meanwhile, we compute the sun position by applying astronomical algorithms which take as input the claimed capture time and location. Only when the two estimated sun positions are consistent can the claimed contextual information be genuine. We have developed a prototype called IMAGEGUARD. The experimental results show that our method can successfully estimate sun position and detect the time-location inconsistency with high accuracy. By setting the thresholds to be 9.4 degrees and 5 degrees for the sun position distance and the altitude angle distance, respectively, our system can correctly identify 91.5% of falsified photos with fake contextual information.
Article
Around the world, metropolitan areas invest in infrastructure for traffic data collection, albeit focusing on highway networks, thus limiting the amount of data available on inner-city roads. For this purpose, various modelling techniques have been developed to interpolate traffic counts spatially and temporally across an entire road network. However, the predictive accuracy of these models depends on the quality and coverage of traffic count data. In this study, we extend the power of spatio-temporal interpolation models with vehicle detection from aerial images, developing a new approach to estimate Annual Average Daily Traffic (AADT) across all roads in an urban area. Using Google aerial images, we extracted the number of vehicles on a road segment and treated these values as observed traffic counts collected over a short period of time. This information was used as input and merged with traffic count data at stations with longer record lengths to predict traffic on all urban roads. This approach was compared against a hold-out sample of roads with observed traffic count data and images, indicating an R-squared (R²) = 90% and RMSE = 7675 between predicted and observed daily traffic counts and R² = 58% and RMSE = 18918 between observed and predicted AADT. The higher prediction accuracy for daily traffic indicates the power of the proposed method for predicting daily values from images; while the lower accuracy of AADT prediction stresses the need for longer-term data to achieve accurate annual averages based on counts derived from images.
Conference Paper
Theoretically, the appearance of shadows in aerial imagery is not desirable for researchers because it leads to errors in object classification and bias in the calculation of indices. In contrast, shadows contain useful geometrical information about the objects blocking the light. Several studies have focused on estimation of building heights in urban areas using the length of shadows. This type of information can be used to predict the population of a region, water demand, etc., in urban areas. With the emergence of unmanned aerial vehicles (UAVs) and the availability of high- to super-high-resolution imagery, the important questions relating to shadows have received more attention. Three-dimensional imagery generated using UAV-based photogrammetric techniques can be very useful, particularly in agricultural applications such as in the development of an empirical equation between biomass or yield and the geometrical information of canopies or crops. However, evaluating the accuracy of the canopy or crop height requires labor-intensive efforts. In contrast, the geometrical relationship between the length of the shadows and the crop or canopy height can be inversely solved using the shadow length measured. In this study, object heights retrieved from UAV point clouds are validated using the geometrical shadow information retrieved from three sets of high-resolution imagery captured by Utah State University’s AggieAir UAV system. These flights were conducted in 2014 and 2015 over a commercial vineyard located in California for the USDA Agricultural Research Service Grape Remote sensing Atmospheric Profile and Evapotranspiration Experiment (GRAPEX) Program. The results showed that, although this approach could be computationally expensive, it is faster than fieldwork and does not require an expensive and accurate instrument such as a real-time kinematic (RTK) GPS.
Chapter
Students’ knowledge of inferential statistics is lacking in many computer science study programs. Yet, the needs for inferential statistical skills have emerged with new fields of study such as human-computer interaction involving observation of human activity. This paper presents experiences teaching inferential statistics to undergraduate computer science students with a focus on the actual goals of the investigations and not the mechanisms and mathematics of statistics. The teaching framework involves teaching statistics as a set of systematic black-box tools.
Chapter
Modelling 3D objects is challenging; often special software skills are required. This paper explores a new method for experimenting with 3D modelling using two-dimensional drawings. These drawings use coloured areas to dictate the rate of curvature. The curvature images are rendered in a radial manner from the centre to the sides. The method allows complex 3D shapes to be modelled. There is no need to employ any new software program as any arbitrary 2D painting application can be used to sketch objects.
Chapter
Many students and researchers struggle with the design and analysis of empirical experiments. Such issue may be caused by lack of knowledge about inferential statistics and suitable software tools. Often, students and researchers conduct experiments without having a complete plan for the entire lifecycle of the process. Difficulties associated with the statistical analysis are often ignored. Consequently, one may end up with data that cannot be easily analyzed. This paper discusses the concept sketch of a framework that intends to help students and researchers to design correct empirical experiments by making sound design decisions early in the research process. The framework consists of an IDE, i.e., Integrated (statistical experiment) Development Environment. This IDE helps the user structures an experiment by giving continuous feedback drawing the experimenter’s attention towards potential problems. The output of the IDE is an experimental structure and data format that can be imported to common statistical packages such as JASP in addition to providing guidance about what tests to use.
Article
Full-text available
This paper presents a methodology to automatically detect and remove the shadows in high-resolution urban aerial images for urban GIS applications. The system includes cast shadow computation, image shadow tracing and detection, and shadow removal. The cast shadow is computed from digital surface model (DSM) and the sun altitudes. Its projection in the pseudo orthogonal image is determined by ray tracing using ADS40 model, DSM and RGB image. In this step, all the cast shadows will be traced to determine if they are visible in the projection image. We used parameter plane transform (PPT) to accelerate the tracing speed. An iterative tracing scheme is proposed. Because of the under precision of the DSM, the fine shadow segmentation is taken on the base of the traced shadow. The DSM itself is short of the details, but the traced shadow gives the primarily correct location in the image. The statistics of the shadow area reflects the intensity distribution approximately. A reference segmentation threshold is obtained by the mean of the shadow area. In the fine segmentation, the segmentation threshold is derived from the histogram of the image and the reference threshold. The shadow removal includes shadow region and partner region labeling, the histogram processing, and intensity mapping. The adjacent shadows are labeled as a region. The corresponding bright region is selected and labeled as its partner. The bright region supplies the reference in the intensity mapping in the removal step.
Article
Full-text available
An adaptive technique for color image segmentation is presented in this paper. The segmentation is performed using a multiresolution scheme and considering the background areas have quite uniform color features at a low-resolution representation of the image. First, a pyramidal representation of the original image is built. Then segmentation is improved iteratively at each resolution using color information. This method allows to extract background areas in outdoor images. Background and foreground separation is especially useful in initialization of object tracking applications. Several color spaces are compared in order to determine a robust method specially with respect to illumination changes which frequently occur in outdoor images. Mean value of Hue component from HSV (Hue, Saturation, Value) color space is selected as the best decision criterion for image matching. Segmentation results of color image from soccer game video sequences are presented to illustrate the method efficiency.
Article
Full-text available
This paper presents a novel algorithm for detecting moving objects from a static background scene that contains shading and shadows using color images. We develop a robust and eeciently computed b ackground subtraction algorithm that is able to cope with local illumination changes, such as shadows and highlights, as well as global illumination changes. The algorithm is based o n a p r oposed c omputational color model which separates the brightness from the chromaticity component. We have applied this method t o r eal image sequences of both indoor and outdoor scenes. The results, which demonstrate the system's performance, and some speed u p t e chniques we employed in our implementation are also shown.
Conference Paper
Full-text available
Automating tools for geo-locating and geo-orienting static cameras is a key step in creating a useful global imaging network from cameras attached to the Internet. We present algorithms for partial camera calibration that rely on access to accurately time-stamped images captured over time from cameras that do not move. To support these algorithms we also offer a method of camera viewpoint change detection, or "tamper detection", which determines if a camera has moved in the challenging case when images are only captured every half hour. These algorithms are tested on a subset of the AMOS (Archive of Many Outdoor Scenes) database, and we present preliminary results that highlight the promise of these approaches.
Conference Paper
Full-text available
Advances in storage capacity means that digital cameras can store huge collections of digital photographs. Typically such images are given non-descriptive filenames names such as a unique identifier, often an integer. Consequently it is time-consuming and difficult to browse and retrieve images from large collections especially on small consumer electronics devices. A strategy for classifying images into geographical regions is presented which allows images to be coarsely sorted into the continent where they were taken. The strategy employs patterns in the time-stamps of images to identify events such as holiday and individual days, and to estimate the approximate longitude where the photographs were taken. Experimental evaluations demonstrate that the continent is correctly estimated for 89% of the images in arbitrary collections and that the longitude is estimated with a mean error of 27.5 degrees. The strategy is relatively straightforward to implement, also in hardware, and computationally inexpensive.
Conference Paper
We extend the triangulation method in [7] and present a tetrahedron method for the sequential mapping of points in high-dimensional space into three-space. The three-space preserves distances among nodes of a tetrahedron. Whenever a new point is mapped, ...
Article
Citation LOYD A. JONES and H. R. CONDIT, "The Brightness Scale of Exterior Scenes and the Computation of Correct Photographic Exposure," J. Opt. Soc. Am. 31, 651-678 (1941) http://www.opticsinfobase.org/josa/abstract.cfm?URI=josa-31-11-651
Conference Paper
A key problem in widely distributed camera networks is geolocating the cameras. This paper considers three scenarios for camera localization: localizing a camera in an unknown environment, adding a new camera in a region with many other cameras, and localizing a camera by finding correlations with satellite imagery. We find that simple summary statistics (the time course of principal component coefficients) are sufficient to geolocate cameras without determining correspondences between cameras or explicitly reasoning about weather in the scene. We present results from a database of images from 538 cameras collected over the course of a year. We find that for cameras that remain stationary and for which we have accurate image timestamps, we can localize most cameras to within 50 miles of the known location. In addition, we demonstrate the use of a distributed camera network in the construction a map of weather conditions.