Viewing Angle Enhanced Point Light Source Display using
Additional Light Sources
Nomin-Erdene Dalkhaa
1
, Gerelmaa Byambatsogt
1
, Densmaa Batbayr
2
and Ganbat Baasantseren
1,*
1
Department of Electronics and Communication Engineering, National University of Mongolia, Ulaanbaatar, Mongolia
2
Ulaanbaatar State University, Ulaanbaatar, Mongolia
Keywords: Displays, Three-Dimensional Display, Image Reconstruction Techniques, Integral Imaging.
Abstract: A Viewing angle (VA) of Point Light Source (PLS) display, which is one type of Integral Imaging (InIm)
display, is low. We proposed to enhance a VA of PLS Display. The new method used nine additional light
sources to increase the VA 2.83 times larger than a conventional method along horizontal and vertical
directions. These additional light sources will increase the angle of the ray that gathered by the elemental
lens. From the experimental results, the VA of the proposed method is 2.55 times larger than the
conventional method.
1 INTRODUCTION
InIm Display is full parallax, full color, natural
depth, and independent of the viewer (B. Lee, 2006).
However, InIm has disadvantages such as short
depth range, narrow viewing angle, and low
resolution (J.-H. Park, 2001), (J.-H. Park, 2005), (S.
-W. Min, 2003).
Three-dimensional (3-D) display images and 3D
view by certain states that angle expressed in terms
of the angle of the range. PLS display determined
using conventional viewing angle is very small.
The previous methods to enhance the VA used
time-multiplexed method (D. -H. Shin, 2006) and
dynamic barrier array (H. Choi, 2003). Those
methods enhanced only horizontal VA and required
different elemental images.
In this paper, we will face according to each
section: the 2
nd
section is viewing angle of the
conventional point light source display, the 3
rd
section is a new method to determine the point light
source display enhancement viewing angle
measures, the 4
th
section is experiment and
discussion, the 5
th
section is conclusions.
2 VIEWING ANGLE OF
CONVENTIONAL POINT
LIGHT SOURCE DISPLAY
The structure of conventional PLS display is shown
in Fig. 1. A light source is at the focal point of the
collimating lens. The elemental lenses collect
parallel rays at the focal points so each elemental
lens creates one PLS. Therefore, it is called PLS
array.
Figure 1: Structure of PLS displays and estimate its
viewing angle.
Rays from PLSs are modulated transmission-
type spatial light modulator (SLM), which is located
behind the lens array, according to an elemental
164
Dalkhaa N., Byambatsogt G., Batbayar D. and Baasantseren G.
Viewing Angle Enhanced Point Light Source Display using Additional Light Sources.
DOI: 10.5220/0006104001640168
In Proceedings of the 5th International Conference on Photonics, Optics and Laser Technology (PHOTOPTICS 2017), pages 164-168
ISBN: 978-989-758-223-3
Copyright
c
 2017 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved

image. Figure 1 illustrates how to create two integral
points. Their y
1,
y
2
are a distance of along Y axis.
Their z
1
, z
2
are a distance of the integrated points P
1
and P
2
from the SLM display. The two integrated
point P
1
and P
2
appear at the cross section of 4 rays,
but VAs of those two points are not equal. However,
the maximum VA of the conventional PLS display is
given by:
.
2
arctan2








⋅
⋅=
f
L
P
C
VA
(1)
where P
L
is a pitch and f is focal length of the
Elemental lens.
The P
1
appears at the cross section of 4 rays.
Two dot-dash lines are not used to create integrated
point P
1
because those two lines are outside
diverging angle of elemental lens L
4
and L
9
,
respectively. The P
2
appears at the cross section of 4
rays. A dot-dash line is not used to create integrated
point P
2
because this one line is outside diverging
angle of elemental lens L
5
respectively.
From the Fig. 1, we can see that the VAs of P
2
and P
1
differ and VA
2
is greater than VA
1
.
Therefore, the VAs of the integrated points are not
the same and depend on the positions of the
integrated point. The VA
1
of P
1
is the sum of α’ and
α. The VA
1
is given by:
.
)5.0'(
arctan
)5.0(
arctan
1












+
−⋅−
+












+
⋅−−
=
=
fz
y
L
Pj
fz
L
Pjy
VA
(2)
where z is the distance according to z-axis, y is the
distance according to y-axis, and j and j’ are the
elemental lens index.
We can conclude that VA of integrated point
increases if the diverging angle of PLS increases.
3 NEW METHOD TO
DETERMINE THE POINT
LIGHT SOURCE DISPLAY
ENHANCEMENT VIEWING
ANGLE MEASURES
We used additional light sources (LS
2
, LS
3
) to
increase the diverging angle of PLS. Figure 2 shows
a structure of the proposed method. The three light
sources are on the focal plane of a collimating lens
and different lateral positions, so light rays of those
3 sources propagate to three different directions.
Those rays are collected by three neighbouring
elemental lenses at the focal point of the center
elemental lens when a distance between two light
sources is equal to s=f'·P
L
/f, where P
L
is a size of the
elemental lens, f' is a focal length of a collimating
lens, f is a focal length of the lens array.
In other word, one elemental lens collects light
rays from the three different sources at the three
positions, as shown in Fig. 2. Hence, one PLS
consists of three parts of up, center, and down
illuminations.
The two additional parts of PLS enhance the
diverging angle of PLS so VA of the proposed
method is enhanced.
Figure 2: Geometrical structure of new method.
From the Fig. 2, we can determine the VA of
proposed method:
.
2
3
arctan2








⋅
⋅
⋅=
f
L
P
P
VA
(3)
According to Eq. (1) and Eq. (3), the VA of
proposed method is larger than the conventional
method.
.
3
)5.0'(
arctan
3
)5.0(
arctan
3












+
−⋅−
+












+
⋅−−
=
=
f
z
y
L
Pj
f
z
L
Pjy
VA
(4)
We can find angles of each integrated point of PLS
display Eq. (4). In other words, we can define VAs
of all the integrated point.
Viewing Angle Enhanced Point Light Source Display using Additional Light Sources
165

4 RESULTS AND DISCUSSION
4.1 Simulation and Discussion
When a focal length of an elemental lens is 3.3 mm
and a pitch of elemental lens is 1 mm, the VA of the
conventional method is 17.2°, according to Eq. (1).
In the same configuration, the VA of proposed
method is 48.8°, according to Eq. (3). It is almost
2.83 times larger than conventional method
theoretically.
Table 1: Parameters of simulation.
Specification Characteristic
Lens array 30 (H) × 30 (V)
Elemental lens
dimension
1 mm (H) × 1 mm (V)
Focal length of lens
array
3.3 mm
Distance between lens
array and SLM new
method
4.4 mm
Parameters of the simulation are listed in Table 1. In
the simulation, the distance of the integrated point is
0-60 mm along the z-axis and 0-30 mm along the y-
axis.
Figure 3: Calculation of the VA. (a) a conventional
method and (b) a new method.
However, PLS is volumetric display, we
calculate the viewing angle on the x=0 yz planes.
PLS is the symmetric display so results on y=0 xz
plane are same as x=0 yz planes.
In Fig. 2, we draw just three light sources, but we
used nine light sources (LS
1
, LS
2
... LS
9
are shown in
Fig. 6) in simulation to enhance VA along the
horizontal and vertical axis. In the simulation, we
calculated the VAs of 180,000 (300×600) integrated
points in both configurations, as shown Fig. 3. The
viewing angles of integrated points at the same
distance from the SLM have small variation, from
the simulation results. In Fig. 3(a), the maximum
VA of the conventional method is 17.2°, according
to Eq. (2). In Fig. 3(b), the maximum VA of the new
method is 48.8°, according to Eq. (4).
We created two sets of elemental images for the
conventional PLS display and a new method, as
shown in Fig. 4.
Figure 4: Elemental images (a) for PLS display and (b)
new method.
The number of elemental images for the new method
is larger than a conventional display. The additional
elemental images of new method enhance VA
because the large diverging angle requires additional
elemental images.
4.2 Experiment and Discussion
The experimental setup is shown in Fig. 6. In the
experiment, we used 1 mm lens array. The
specification of the lens array is given in Table 1.
Therefore, in the experiment, we only used the
two InIms that are discussed in the simulation. In
each experiment, we took pictures after we changed
the position of the camera and then rotated the
camera when it was within the boundary between
the viewing regions of the InIm.
The camera is similar to a viewer, so the angle of
the camera is the VA of the integrated point. In the
experiment, we used two objects. Distances of
objects, “A” and “T”, are 10 mm and 20 mm from
the lens array, respectably, as shown in Fig. 5.
PHOTOPTICS 2017 - 5th International Conference on Photonics, Optics and Laser Technology
166

Figure 5: Experimental configuration.
Figure 6: Set-up of the proposed method.
We used 9 (3×3) Light-Emitting Diode (LED)
instead of the light sources to enhance the VA along
horizontal and vertical direction, as shown in Fig. 6.
Figure 7(a) shows the experimental result of PLS
when the just one source LS
1
is turned on. It is like
the conventional PLS. When up source LS
2
, center
sources LS
1
, and down source LS
3
are turned on and
other LEDs are turned off, the results is shown in
Fig. 7(b).
Therefore, PLSs in the middle are brighter and
bigger than other PLSs. Figure 7(c) shows results
when left source LS
4
, center source LS
1
, and right
source LS
5
are turned on and other LEDs are turned
off.
From the Figs. 7(b) and 7(c), the elemental lens,
where in the center of PLS array, can collect rays
from additional light sources, the lens array must be
close to the collimating lens.
Therefore, we changed the distance between the
lens array and the collimating lens from 50 mm to 5
mm. The PLSs are shown in Fig. 7(d). From Figs.
7(c) and 7(d), brighter part of PLSs are large when
the lens array is close the collimating lens.
Figure 7: PLS displays in front view (a) the LS
1
LED
turned on (b) the LS
2
, LS
1
, LS
3
LEDs are turned on (c) the
LS
4
, LS
1
, LS
5
LEDs are turned on with l by 50 mm and (d)
the LS
4
, LS
1
, LS
5
LEDs are turned on with l by 5 mm.
We took the pictures when the new method
displays two objects where “A” and “T” are 10 mm
and 20 mm from SLM, respectably. In the
experiment, we displayed two InIms that are on the
center view as shown in Fig. 8(c).
Figure 8: PLS displays of front view (a) left 18° (b) right
18° (c) center 0° (d) left 22° and (e) right 22°.
When we moved the camera from left to right,
we took pictures in five different positions and
measured the angle of the camera. After the invisible
object “T”, the angle of the camera from the right
side is 18°. Before the invisible object “A”, the angle
of the camera from the left side is 18°.
Before the invisible objects “T” and “A”, the
angle of the camera from the left side is 22° and
Viewing Angle Enhanced Point Light Source Display using Additional Light Sources
167

from the right side is 22°. Therefore, the VA of the
objects “T” and “A” are 44°. From the experimental
result, the VA is 2.55 times larger than the
conventional method.
5 CONCLUSIONS
We use nine light sources to increase the diverging
angle of the PLS along both of horizontal and
vertical axis. The large diverging angle enhanced the
viewing angle. When the distance between two light
sources is given by Eq. equal to s=f'·P
L
/f, viewing
angle of proposed method is 2.55 times larger than
conventional PLS display. New method adds just
light sources. It is different from other methods.
REFERENCES
Lippmann, 1908. La Photographie Integrale, Comptes-
Rendus Academie des Sciences 146, 446.
B. Lee, J. -H. Park, and S.- W. Min, 2006. "Three-
dimensional display and information processing based
on integral imaging," in Digital Holography and
Three-Dimensional Display, T.-C. Poon, eds.
(Springer, New York, USA).
J.-H. Park, S. -W. Min, S. Jung, and B. Lee, 2001.
"Analysis of viewing parameters for two display
methods based on integral photography," Appl. Opt
40, 5217-5232.
J. -H. Park, J. Kim, J.- P. Bae, Y. Kim, and B. Lee, 2005.
"Viewing Angle Enhancement of Three-
Dimension/TwoDimension Convertible Integral
Imaging Display Using Double Collimated or
Noncollimated Illumination," Jpn. J. Appl. Phys 44,
991- 994.
S. -W. Min, B. Javidi, and B. Lee, 2003. "Enhanced three-
dimensional integral imaging system by use of double
display devices, " Appl. Opt, 42, 4186-4195.
D. -H. Shin, B. Lee, and E. -S. Kim, 2006.
"Multidirectional curved integral imaging with large
depth by additional use of a large-aperture lens," Appl.
Opt, 45, 7375-7381.
H. Choi, S. –W. Min, S. Jung, J. –H. Park, and B. Lee,
2003. "Multiple-viewing-zone integral imaging using
a dynamic barrier array for three-dimensional displays,
" Appl. Opt 11, 927-932.
PHOTOPTICS 2017 - 5th International Conference on Photonics, Optics and Laser Technology
168