Paper Presentation on 'Pedestrian Detection Using Night Vision Technology'

1. 1
PEDESTRIAN DETECTION USING NIGHT VISION TECHNOLOGY
1Pavithra.B, II Year, Department of IT
2J.Kalaivani, Associate Professor, Department of IT
IFET College of Engineering,Villupuram, TamilNadu, India
Abstract— The term ‘night vision’ literally means ‘to see in
low-light conditions’. A night vision system increases
visibility in situations where only low beam head lights are
being used. Night Vision Technology is being used by federal
and state agencies for security, surveillance and search and
rescue operations. Pedestrian detection is one of the fields of
research for both government and commercial organizations.
Road accidents are frequent nowadays and death rates due to
them are increasing day by day. Night Vision Technology is
used to rectify it by installing a night vision system in the
automobile to track the pedestrians at the night and to avoid
collisions with them. It is more advantageous when compared
to the existing systems. This technology would pave way for
self driving vehicles.
I. INTRODUCTION
It is very difficult to see in the dark, especially for us
humans when compared to other animals. We cannot adapt to
a darker environment all of a sudden. In darkness, the eye sees
only a fraction of what it sees in daylight. Our eyes may need
more time to adjust from a brightly lit space into a dark space,
such as a movie theatre. Moreover, most of the people suffer
from night blindness, a below-average ability to see at night or
in low light. Such conditions can occur in people of all ages,
even young children. People with night blindness often have
problems driving at night.
According to the National Safety Council, traffic death
rates are three times greater at night than during the day. A
driver’s ability to react depends on vision, and normal vision
is severely limited at night. Driving at night is dangerous, and
that too for the older citizens it is a great burden because they
experience more difficulty in driving at night than others. Here
comes the Night Vision Technology to save the day.
The term ‘night vision’ literally means ‘to see in low-light
conditions’. So, by using the appropriate night-vision
equipment one could see a person standing 200 yards away
that too on a moonless, cloudy night without any strain! Thus
comes the use of night vision systems which uses infra-red
sensors or headlights to provide a clear view of the road ahead
and helps in detecting the pedestrians or animals on the road.
A night vision system increases visibility in situations where
only low beam head lights are being used.
II.HISTORY OF NIGHT VISION
Earlier, man regarded the night time to be the most
dangerous and did all his work at daytime. The discovery of
fire helped him to lower his fears about the dark. But, still he
was afraid of the dark. Slowly, he began to hunt at nighttime
with the help of a fire torch. This was not a valid option since
the fire may go out at anytime and he needs to relit it in the
dark. And then, man realized that night is very beneficial for
him to attack his enemies. Using torches proved to be a fatal
approach as it revealed the position of the attacker. A solution
was provided for this issue by the innovation of night vision in
the 1940’s during the World War. Initially, this technology
was only used in sniper rifles to foresee enemies from a long
distance. Germans were the first to develop the night vision
technology so they could dominate their enemies at night.
During the 1960’s Vietnam Era, the evolution of this
technology by the military troops led to the invention of night
vision goggles. Night Vision Technology was the prime factor
used during the war against Iraq. It became an excellent tool
for defensive purposes and guarding against the militia. It has
provided the United States with a strategic military advantage
over other nations. It was also used by hunters as it is easy to
track animals at night.
At present, Night Vision Technology is being used by
federal and state agencies for security, surveillance and search
and rescue operations. Civilians can also make use of this
technology through proper night vision equipment. Even
investigators and law enforcement are using this technology to
nab criminals. Bulky night vision equipments were replaced
with affordable and lightweight products with enhancement.
III.GENERATIONSOF NIGHT VISION
The generations of night vision are given below:
1. GENERATION 0
Earlier, instead of intensification, image conversion
was used in night vision products. They required a
source of invisible infrared light to illuminate the
target area.
2. GENERATION 1
In this generation, three image intensifier tubes are
connected in series. The systems are larger and
heavier.
3. GENERATION 2
Image Intensification was used which increased the
gain and resolution by employing a micro channel
plate.

2. 2
4. GENERATION 3
Fig. 1. Generation 3 system
The GaAs photocathode and the ion barrier film was enabled
on the micro channel plate which proved to be the major
advancements in this generation.
5. GENERATION 4
‘Filmless and Gated’ image intensifiers were
provided which not only increased the detection range in
pitch-black conditions but also reduced the halo which
responded to bright sources.
IV.TECHNOLOGIESOF NIGHT VISION
FOLLOWINGARE THETECHNOLOGIESOF NIGHT VISION:
1. Thermal Imaging
This technology operates by capturing the upper
portion of the infrared light spectrum, which is
emitted as heat by objects instead of simply reflected
as light. Hotter objects, such as warm bodies, emit
more of this light than cooler objects like trees or
buildings.
In this method, a special lens focuses the
infrared light emitted by all of the objects in view.
The focused light is scanned by a phased array of
infrared-detector elements. The detector elements
create a very detailed temperature pattern called a
thermogram. The detector array obtains the
temperature information to make the thermogram.
The thermogram created by the detector elements is
translated into electric impulses. Such impulses are
sent to a signal-processing unit, a circuit board with a
dedicated chip that translates the information from
the elements into data for the display. The signal-
processing unit sends the information to the display,
where it appears as various colors depending on the
intensity of the infrared emission. The combination
of all the impulses from all of the elements creates
the image.
2. Image Enhancement
This works by collecting the tiny amounts of
light, including the lower portion of the infrared light
spectrum, that are present but may be imperceptible
to the naked eye, and amplifying it to the point which
can be observed easily.
In this method, a conventional lens, called the objective lens
captures ambient light and some near-infrared light. The
gathered light is sent to the image-intensifier tube. The image-
intensifier tube has a photo cathode, which is used to convert
the photons of light energy into electrons. As the electrons
pass through the tube, similar electrons are released from
atoms in the tube, multiplying the original number of electrons
by a factor of thousands through the use of a micro channel
plate (MCP) in the tube. At the end of the image-intensifier
tube, the electrons hit a screen coated with phosphors. These
electrons maintain their position in relation to the channel they
passed through, which provides a perfect image since the
electrons stay in the same alignment as the original photons.
The energy of the electrons causes the phosphors to reach an
excited state and release photons. These phosphors create the
green image on the screen that has come to characterize night
vision. The green phosphor image is viewed through another
lens, called the ocular lens, which allows you to magnify and
focus the image. The NVD may be connected to an electronic
display, such as a monitor, or the image may be viewed
directly through the ocular lens.
Night vision can be categorized as:
1. Biological night vision
The molecules of rhodopsin in the rods of the eye
undergo a change when the light is absorbed by them.
2. Technical night vision
It makes use of night vision device(NVD), an optical
device that allows images to be produced in levels of
light approaching total darkness.
Before moving on, we should be aware of certain terms
associated with light. As we all know, light is made up of
finely divided particles. The amount of energy in a light wave
is related to its wavelength: Shorter wavelengths have higher
energy. Of visible light, violet has the most energy, and red
has the least. Just next to the visible light spectrum is the
infrared spectrum.
Fig. 2 Spectrum of Light
Infrared light can be split in three ways:
1. Near infrared: It is the most closest to the visible
light.
2. Mid infrared: It ranges from 1 to 1.3 microns.
3. Thermal infrared: It occupies the largest part of the
infrared spectrum.
V.PEDESTRIAN DETECTION
Pedestrian detection is one of the field of research for both
government and commercial organizations. Humans don’t
have an effective eye sight in low light compared to many
animals due lack of Tapetum Lucidum (a layer of tissues in
the eye of many vertebrates, which lies immediately behind
retina). Night vision acts as a tool to see in the complete dark

3. 3
condition. The night vision system provides black and white
images of the driving environment in the information display
called as Central Information Display (CID). Objects situated
ahead of the vehicles are shown with the varying degrees of
brightness depending on the respective temperatures.
The night vision system consists of four main modules:
a) Input Video: We take the available datasets, in the
input video that are captured using NIR camera.
The pedestrian generally appears uniform and
slightly brighter than the background and other
objects in the images.
b) Frames: From the input video, frames ( the
electronically coded still images) are extracted. In
the video we can get better sense of motion by
increment in the frame rate. The standard expected
frame rate is to be 25 frames per second and
29.97fps depends on the video.
c) Segmentation: It refers to the process of
partitioning a digital image into multiple segments.
In a simple way segmentation is nothing but
divides and analyze. Output of image segmentation
is a set of segments that completely covers the
entire image. Each pixel in a region is similar to
the computed property, characteristics, color,
intensity, or texture.
d) Edge Detection: By applying this methodology,
the quantity of processed data will be reduced
retaining important information regarding shapes
of pedestrians. It consists of smoothing processed
images with reduction in noise with the help of
filters like Gaussian filter which works without
eliminating the true edges of the images.
e) Detection of blob: The out of the image from
vertical edge detector is passed to this section
where merging is done for block identification. It
will identify the pedestrian like objects in the
images like images blocks that differs with the
properties like brightness, color compared to the
surrounding regions in the images.
f) Merging of blob: It is necessary because if the
pedestrian is close to the camera or vehicle is
moving edges of the same pedestrian are separated
by distance. In such cases the blobs are separated.
As a result, there is a probability of losing
pedestrians, because the individual blobs may not
singly satisfy the height and width criteria also
relabeling the pixels.
g) Tracking: The motion of the vehicle or moment of
pedestrians or because of both the pedestrian
position varies frame to frame. Visual surveillance
application commonly uses pedestrians tracking
system. Tracking aims to understand behavior and
activity recognition. A searching window is a
formed by enlargement of the height and width of
the block that is selected from previous frame in
either vertical or horizontal direction used for
tracking. Tracking can be done in two different
ways either template matching or by segmentation
process.
Fig. 3. Night Vision System
The below figure represents the block diagram of a night
vision system.
Fig.4. Block Diagram of Algorithm
VI. IMPLEMENTATION
The night vision systemmentioned above can be implemented
in the automobile industry as follows.
Fig. 5.Components of a night vision system
i. Night Vision Camera: The thermal imaging camera
consists of a heated optical element and thermal

4. 4
imaging sensor in it. The thermal imaging sensor is
made up of a multiple sensor elements. Each display
pixel is assigned one sort of sensor element. The
sensor elements generate an electrical signal as per
the intensity of heat radiation. The higher the
temperature, the brighter the corresponding pixel
will be displayed in it.
Fig. 6. Night vision camera
ii. Night Vision Control Unit: The accommodation of
control unit is in the front device holder behind the
glove box. The diagnosis of data, programming and
coding data are transmitted to the camera with the
help of the control unit. The camera and the front-
lens heater are powered via the control unit. In
addition to this, the control unit converts the
symmetrical image data from the camera to the
CVBS signal and, depending on the equipment
specification making the signal available to either
the navigation systemor the video module.
iii. Button in Light Switch Center: The button for
switching Night Vision on or off is provided in the
light switch center. The followed condition can
exists like: The rain/light sensor detects the
sufficient ambient light and driving lights are
switched off; BMW night vision is ready for
operation approx. 2 seconds after the button in the
light switch centre is pressed for application.
iv. Sensor: The rain/light sensor detects insufficient
ambient light and driving lights are switched on
back to original state.
Fig.7. Switch button
VII. PRINCIPLE OF OPERATION
The Night Vision camera converts thermal radiation into
electronic signals and then into images visible to the human
eye. The thermal image is converted first by the sensor into
electrical signals and then with the aid of image-processing
software into a visible image in the control display. The sensor
elements alter the resistance in proportion to the temperature.
The higher the temperature, the higher the electrical signal and
the whiter the pixel will be shown. The sensor can generate a
new image up to 60 times per second. This results in a softer
and clearer image. Heat radiation is absorbed and dissipated
by virtually every solid or liquid body. Heat radiation,
however, is not visible to the human eye because it belongs in
the long-wave infrared range. From a physical standpoint, this
represents electromagnetic waves with a wavelength of 8 μm
to 15 μm. This long-wave infrared radiation is known as Far
Infrared (FIR). The advantage of utilizing radiation in the Far
Infrared range is the greater range compared with Near
Infrared systems with a wavelength of 0.7 μm to 1.4 μm.
These systems require illumination with just this wavelength.
Essentially, FIR systems consist of an optical element, a
thermal imaging camera, a control unit and a display.
Fig.8. Principle of Operation
VIII. ADVANTAGES OVER PREVAILING SYSTEMS
Night vision technology has proved to be a boon in many
ways. Some of them are:
1. It has improved the vision conditions of dusk and
darkness.
2. There is no dazzling by head lights of the oncoming
vehicles.
3. It ensures the highlighting of illuminated, heat-
emitting objects as pedestrian, cyclists, deer, etc.
4. It gives a better overview of the driving situations.
5. The zoom functions of the object in the far distance
at high speeds.
6. It illuminates the bends/curves.
7. It provides illumination of dark courtyard and garage
entrances.
8. It has superior image quality.
9. It provides an uninterrupted image.
10. It is immune to dynamic thermal environment.
IX. FUTURE ASPECTS
Considering the huge list of benefits gained through the night
vision technology over other systems, it could be predicted

5. 5
that in near future, the advancement of this technology along
with the GPS(Global Positioning System) Navigation will
bring about the advent of driverless vehicle – an automobile
that self drives without any aid of the owner. Effective night
vision could be the key parameter in such vehicles. This
technology will probably be applied in hidden object detection
and wildlife observations. There would be a time when the
world comes alive even after dusk, and no one will be afraid
of the dark as our forefathers.
X. CONCLUSION
Night Vision Technology is one of the emerging domains.
Although it has various applications, its advent in the
automobile industry has provoked a huge response from the
civilians. It could be a boon to avoid future accidents and thus
could be a life saver. The main objective of this technology in
the vehicle is to avoid the road accidents. On the basis of that
we conclude that automatic pedestrian warning, in the formof
highlighting the pedestrians on the night vision display, is
generally helpful in increasing detection distance and
accuracy. Overall, this paper summarizes a variety of qualities
and necessities of night vision technology. A new approach
has been developed to detect pedestrian using technique at
night time with low light or no light. The success of the
proposed implementation of this system opens new frontiers
for further research in the future time. Future work can be
included with a classifier to decrease the number of false
positives occurred for better perfection and safety.
REFERENCES
[1] A Simple and effective Display foe night vision
system, Omer Tsimhoni ,Michael J. flannagan, Mary
Lynn Meford and Naoko Takenobu, University of
Michigan transportation research Institute
[2] Pedestrian detection with night vision system
enhanced by automatic warnings by Omer Tsimhoni,
Mchael J. Flannagan and Takako Minoda.,
September, 2005
[3] International Journal of Environmental Science
Volume-3, No.1, 2012
[4] Improved Night-vision for next 2009 BMW 7 –series
Bmwblog 2008-12-31.Retrieved 2009-12-08
[5] Night vision: requirements and possible roadmap for
FIR and NI System
[6] S. Ahirerao, "Night Vision System in BMW ",
International Review of Applied Engineering
Research, ISSN 2248-9967 Volume 4, Number 1
2014, pp. 1-10.
[7] R. P. Raghatate, S. S. Rajurkar, M. P. Waghmare, P.
V. Ambatkar, "Night Vision Techniques and Their
Applications", International Journal of Modern
Engineering Research, ISSN: 2249-6645, Vol.3,
Issue.2. March-April. 2013, pp. 816-820.