VISVESVARAYA TECHNOLOGICAL UNIVERSITY

Jnana Sangama, Belgavi-590018

A Technical Seminar Report

on
NIGHT VISION TECHNOLOGY
Submitted in partial fulfillment of the requirements for the Degree of

Bachelor of Engineering
In
Electronics and Communication Engineering
Submitted by

CHITHRA K 1SJ16EC121

Under the guidance of Seminar Coordintors

Mr. Ravikiran R Mr. Mohan Babu C & Ms. BhavanaS
Asst. Professor Asst. Professors
Dept. of ECE, SJCIT Dept. of ECE, SJCIT

S J C INSTITUTE OF TECHNOLOGY
DEPARTMENT OF ELECTRONICS AND COMMUNICATION ENGINEERING
CHICKBALLAPUR-562101
2019-2020
 S J C INSTITUTE OF TECHNOLOGY
CHICKBALLAPUR-562101
(Affiliated to Visvesvaraya Technological University, Belagavi, Karnataka)

DEPARTMENT OF ELECTRONICS AND COMMUNICATION ENGINEERING

CERTIFICATE

This is to certified that CHITHRA K bearing USN 1SJ16EC121 has satisfactorily completed
technical seminar entitled “NIGHT VISION TECHNOLOGY” in partial fulfillment of the
requirements as prescribed by the Visvesvaraya Technological University, Belagavi for VIII
semester, Bachelor of Engineering in Electronics and Communication Engineering and submitted
this report during academic year 2019-2020.

-------------------------------- ------------------------------------ -------------------------------

Signature of the Guide Signature of Seminar coordinators Signature of the HOD
Mr. Ravikiran R Mr. Mohan Babu C and Ms. Bhavana S Dr. B.N. Shobha
Asst. Professor Asst. Professors HOD
Dept. of ECE, SJCIT Dept. of ECE, SJCIT Dept. of ECE, SJCIT
 ACKNOWLEDGEMENT

The satisfaction and euphoria that accompany the completion of any task would be
incomplete without the mention of the people who made it possible, whose constant guidance and
encouragement ground my efforts with success.

I consider it is a privilege to express my gratitude and respect to all those who guided me in
completion of the technical seminar report.

I submit my humble pranamas to MAHASWAMIJI, for providing all the required facilities
and the conductive environment in the campus.

I am grateful to our beloved Principal Dr. K. M. RAVIKUMAR for his inspiration towards
my academic progress and for providing all the required facilities and the conductive
environment in the campus.

It is a great privilege to place on record my deep sense of gratitude to Dr. B. N. SHOBHA,

Professor and Head, Department of Electronics and Communication, who patronized throughout
my career and for the facilities provided to carry out this work successfully.

I would like to thank seminar co-ordinators Prof. MOHAN BABU. C and Prof.
BHAVANA. S and my technical seminar guide Prof. RAVIKIRAN. R Department of
Electronics and Communication for their valuable support and guidance.

I thank the teaching and non-teaching staff members who have helped me directly or
indirectly during the technical seminar work.

Finally, I would like to thank my family members and friends for their co-operation and
motivation to complete this technical seminar work successfully.

CHITHRA K (1SJ16EC121)

i
 LIST OF FIGURES

Figure no. Name of figures Page no.

Fig 2.1 The basic components of thermal imaging system 4

Fig 2.2 Image intensifier tube 5

Fig 3.1 Apple image during day and night 6

Fig 3.2 Gray image during day and night 7

Fig 3.3 Classification of image denoising 7

Fig 3.4 The figure of ICA noisy model 8

Fig 4.1 The pure picture added with noise 10

Fig 4.2 The median-mean filtering denoising method 10

Fig 4.3 The FastICA denoising method 11

Fig 4.4 The improved fastICA algorithm denoising method 11

ii
 ABSTRACT
Night vision is referenced as technology that provides us with the miracle of vision in total
darkness and improvement of vision in low light environments. The technology is an amalgam of
several different methods each having its own advantages and disadvantages. The most comman
methods described here are Low-Light imaging, Thermal Imaging and Illuminations. And also
gives brief idea about night vision devices, Night Vision Image denoising methods. Humans have
poor night vision compared to many animals, in part because the human eye lacks a tapetum
lucidum. A night vision device (NVD) is an optical installment that allows images to be produced
in levels of light approaching total darkness. They are most often used by the military and law
enforcement agencies, but are available to civilian users. The term usually refers to a complete
unit, including an image intensifier tube, a protective and generally water resistant housing, and
some type of mounting system. Many NVDs also include sacrificial lenses, IR illuminators, and
telescopic lenses. NVDs are mounted appropriately for their specific purpose, with more general-
purpose devices having more mounting options.

iii
 TABLE OF CONTENTS
Page No

ACKNOWLEDGEMENT i

LIST OF FIGURES ii

ABSTRACT iii

CHAPTER 1 PREAMBLE 1
1.1 Introduction

1.2 Objective

1.3 Literature Survey

CHAPTER 2 BRIEF DISCRIPTION OF

THETOPIC 4
2.1 Thermal Imaging
2.2 Image Enhancement

CHAPTER 3 NIGHT VISION IMAGE

DENOISING 6
3.1 Noise Analysis
3.2 Denoising Methods
3.2.1 ICA
3.2.2 FastICA
3.2.3 Maximum Likelihood deviation
Removed FastICA

CHAPTER 4 PERFORMANCE ANALYSIS 10

CHAPTER 5 ADVANTAGES AND
APPLICATIONS 12
5.1 Advantages
5.2 Applications
CHAPTER 6 CONCLUSION 13
REFERENCES 14
Night Vision Technology

CHAPTER 1
PREAMBLE

1.1 INTRODUCTION
Night vision signifies the ability to see in dark (night). This capability is normally
possessed by owls and cats, but with the development of science and technology devices has
been develop which enables human being to see in dark as well an in adverse atmospheric
conditions such as fog, rain, dust etc. The muscles in the human eye have the ability to stretch or
contract automatically, depending upon the intensity of light falling on the eye. When we go out
in bright sunlight, the pupil gets contracted. Alternatively, when we enter a shaded or dark room
at that time the muscles of eye relax and make the aperture of the eye lens big enough to allow
sufficient amount of light to pass through, therefore the objects in the room appear blurred.
Because of this human eye have limitations. The muscles of eye cannot increase the aperture
indefinitely. Therefore, in poor light we are unable to see the objects because the image cannot
be formed on the retina clearly. The capability to detect and identify targets at night and under
poor visibility conditions has been an essential military requirement[1]. The modern army's need
to operate at night and under conditions of extremely poor visibility , Since the soldiers have to
often fight in the dark at night, they have to face a severe stress as far as the location of target is
concerned. Also various wild life observer have to face problems of low light because many wild
animals are more active during night time than day, therefore to observe there lifestyle and study
it night vision is important. Therefore to make human being unable to see in dark by
technological means, night vision technology has been developed[2].

Night vision technologies can be broadly divided into three main categories:

1. Image intensification
2. Active illumination
3. Thermal illumination

Image intensification: This method basically involves ambient light amplification. It works by
being able to detect low levels of light and then amplify it. When photons (tiny energy packets
that make up light) enter an image enhancer they first hit a layer called a photo cathode which

Dept. of ECE, SJCIT Page 1 2019-2020

Night Vision Technology

releases electrons. These electrons hit a second layer called a micro channel plate which
multiplies the electrons before they hit the phosphor screen which converts them back into light
because there are now so many more electrons that we get a brighter image.
But this model fails if there isn’t enough light for the thermal enhancers to see at all[2].

Active illumination: Active illumination technologies work on the principle of coupling imaging
intensification with an active source of illumination in the near infrared (NIR) band[2] Infrared is
used in night vision technology when there is insufficient visible light to see, active illumination
involves conversion of ambient light photons into electrons which are then amplified by a
chemical and electrical process and then converted back into visible light.

Thermal illumination: A lens focuses the infrared light this light is scanned and this creates a
temperature pattern. Now the generated pattern is translated into electric impulses then sent to a
circuit board which translates it into data for the display[3].Thermal images are black and white
in nature.

1.2 OBJECTIVE
To know about different night vision technologies and night vision image denoising
methods.
1.3 LITERATURE SURVEY

P.R. Namitha and et.al [1] A night vision system must increase visibility in situations where
only low beam head lights used today. As pedestrians and animals have highest risk increase in
night time traffic due to darkness, the ability of detecting those objects should be the main. In the
night vision assistant driving systems with near IR active imaging, the contrast and resolution of
images or videos captured by cameras are drastically altered by back scattering in bad weather
such as fog and mist etc, which brings a hidden danger in. Monitoring crime on the highway is
crucial process. Vehicle Detection and Tracking System from CCTV captured image for night
vision. In night time, it is very difficult to identify the vehicles due to unwanted light sources. In
this paper we present a method to eliminate.

Dept. of ECE, SJCIT Page 2 2019-2020

Night Vision Technology

Mohd Junedul Haquel and et.al [2] Image Processing refers Capturing and manipulating
images to enhance or extract information. Image processing is a form of signal processing for
which the input is an image, such as a photograph or frame. The output of image processing may
be either an image or, a set of characteristics or parameters related to the image. This paper is
about Night vision Technology, by definition, literally allows one to see in the dark, originally
developed for military use. Night vision can work in two very different ways, depending on the
technology used. Image enhancement–This works by using the lower portion of the infrared light
spectrum. Thermal imaging - This technology operates by using the upper portion of the infrared
light spectrum.

Rupesh P.Raghatate and et.al [3] This paper describes the various Night vision
techniques."Night Vision" is referenced as technology that provides us with the miracle of vision
in total darkness and the improvement of vision in low light environments. This technology is an
amalgam of several different methods each having its own advantages and disadvantages. The
most common methods described here are Low-Light Imaging, Thermal Imaging and
Illumination’s. This paper also give brief idea about various night vision device (NVD) that
allows images to be produced in levels of light approaching total darkness, it also explains
various applications where night vision technology is used to solve various problems due to low
light conditions.
Hongjun Wang and et.al [4] When apple harvesting robot operates at night, there are lots of
noise in the apple’s night vision image captured by image processing system. And the noise is
mainly Gaussian noise, and mixed with some Salt and Pepper noise. In order to improve the
harvesting efficiency and precision, the Independent Component Analysis (ICA) is introduced
into the denoising method for night vision image. Mixture ICA model with noise based on the
Maximum Likelihood Estimation regards likelihood of the observed signals as the objective
function, which leads to the estimation to the mixed matrix poor. Aiming at the problem, this
paper improves the objective function by bias removal technique, which can reduce the bias
caused by the noise.

Dept. of ECE, SJCIT Page 3 2019-2020

Night Vision Technology

CHAPTER 2

BRIEF DISCRIPTION OF THE TOPIC

Night vision technology can work in different ways based on technology used. Now let
us discuss about how image enhancement and thermal imaging works[9].

2.1 THERMAL IMAGING

 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. It only
takes about one-thirtieth of a second for the detector array to obtain the temperature
information to make the thermogram. This information is obtained from several thousand
points in the field of view of the detector array.
 The thermogram created by the detector elements is translated into electric impulses.
 The 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.

Fig 2.1: The basic components of thermal imaging system

Dept. of ECE, SJCIT Page 4 2019-2020

Night Vision Technology

2.2 IMAGE ENHANCEMENT

Image-enhancement technology is what most people think of when you talk about night
vision. In fact, image-enhancement systems are normally called night-vision devices (NVDs).
NVDs rely on a special tube, called an image-intensifier tube, to collect and amplify infrared and
visible light. Here is how it works[3]

 The conventional lens called the objective lens captures ambient light and some near –
infrared light.
 The image- intensifier tube has a photo cathode, which is used to convert the photons of
light energy into electrons.
 At the end of the image- intensifier tube, the electrons hit a screen coated with phosphors.
 The green phosphors image is viewed through another lens, called the ocular Lens which
allows you to magnify and focus the image. The LED may be connected to an electronic
display, such as a monitor, all the image may be viewed directly through the ocular lens.

Fig 2.2: Image intensifier tube

Dept. of ECE, SJCIT Page 5 2019-2020

Night Vision Technology

CHAPTER 3

Night vision image de-noising method

Digital image is inclined to a variety of noise which affects the quality of image.
Unwanted electrical fluctuations themselves came to be known as "noise. The main purpose of
de-noising the image is to restore the detail of original image as much as possible. The criteria of
the noise removal problem depends on the noise type by which the image is being corrupted. In
the field of reducing the image noise several type of linear and non linear filtering techniques
have been proposed[4].Salt and pepper noise is a form of noise typically seen on images. It
represents itself as randomly occurring white and black pixels. Principal sources of Gaussian
noise in digital images arise during acquisition eg., sensor noise caused by poor illumination
and/or high temperature. The values that the noise can take on are Gaussian-distributed.
However, in many physical environments, the noise exhibits impulsive characteristics, which
cannot be adequately described by a Gaussian model. We have many de-noising techniques like
Median filtering(MF), Wavelet de-noising method(WT), A simple fuzzy method, Independent
component analysis(ICA), Fast ICA (Fixed point algorithm) and so on[10].

3.1 Noise Analysis

Now let us take an example of apple harvesting robot operates at night for noise analysis,
Compared with natural light, on one hand, artificial light source has the shortages of small range
of light, uneven light, poor stability etc, on the other hand, low temperature and high humidity at
night also affect the image quality to a certain extent collected by robot image processing system.
Here, apple picking robot is taken as the research object, mainly carries on image processing to
apple images collected at night[4]. The images during day and at night are shown in fig 3.1.

Fig 3.1: Apple image during day and night

Dept. of ECE, SJCIT Page 6 2019-2020

Night Vision Technology

Color image contains a large amount of information, processing speed is slow, and ICA
algorithm requires iterative operation, so in order to avoid the program running too slow,
transforming the images into gray scale first, as shown in Fig 3.2.

Fig 3.2: Gray image during day and night

It can be seen from Fig 3.2 that the night vision apple image is seriously polluted by noise, some
of which can be seen as Salt and Pepper noise through the visual judgment, the other part can not
be visually judged. In the past, the noise analysis of night vision images is dominated by
Gaussian noise. Now we apply different de-noising algorithms to get noiseless images.

3.2 De-noising methods

Image Denoising

Spatial Domain Transform Domain

Non Linear Filter Linear Filter Non Adaptive Data Adaptive

data transform Transform

Median Mean Spatial frequency

domain ICA
Weiner
Spatial Median
Wavelet domain

Weighted Median

Linear Filtering Non-Linear Wavelet coefficient Non-Orthogonal

Threshold Filtering model Wavelet transform

Fig 3.3: Classification of image denoising[10]

Dept. of ECE, SJCIT Page 7 2019-2020

Night Vision Technology

3.2.1 ICA -Mathematical model

The noisy model of ICA:

x(t)=As(t)+n(t) (1)

Where x(t)=[x1(t),x2(t),……..,xM(t)]T means M observation signals, s(t)=[s1(t),s2(t),….,sN(t)]T

means N independent source signals of observed signals, n(t)=[n1(t),n2(t),……,nM(t)]T means
additive noise of M dimension and the mixed matrix A=[a1,a2,......,aN]€ RM×N (M×N) is full rank,
ai is the base vector of A. So it can be seen that the observation signal is weighted by the mixed
matrix. Since A is unknown, assuming that the separation results are independent of each other,
the key to obtaining the source signal is to find a separation matrix W ,which can causes the
output signal y(t) to approach the source signal s(t) as much as possible[4]. The estimated signal
is indicated as (2)

y(t)=W x(t) (2)

Source Signals Noise Observed Signals Estimated signals

S1(t) X1(t) Y1(t)

S2(t) A X2(t) W Y2(t)

SM(t) XM(t) YM(t)

Mixed system Separated system

Fig 3.4: The figure of ICA noisy model

3.2.2 FastICA

At present, it is still a problem to denoise the Gaussian noise and mixed noise. Although
many scholars put forward a lot of methods, the final effect is still not ideal. Based on the
principle of non-Gaussian maximization, FastICA uses the Fixed-point algorithm’s iterative
theory to find the max non Gaussian maximum value of WTx. The algorithm uses Newton
iterative algorithm to handle a large number of sampling points of the observed signal in batch.

Dept. of ECE, SJCIT Page 8 2019-2020

Night Vision Technology

As the parallel processing, simple calculation, small required memory with good convergence of
the algorithm, it can reduce the running time for the image denoising process[4].

The basic steps of FastICA algorithm:

Step 1: Center and whiten the observation signal x, and the matrices D and H are obtained
respectively;

Step 2: Randomly select the initial weight vector, where W0, k=0 ;

Step 3: Update the weight vector Wi+ by iterative formula;

Step 4: Normalize Wi+, namely Wi+ /||Wi+||Wi+;

Step 5: If |Wi+1–Wi|>ε means that the algorithm does not converge, then return to step (3);
otherwise FastICA estimates an independent component, and the algorithm ends.

3.2.3 Maximum Likelihood deviation Removed FastICA

The main algorithms of noisy model of ICA include deviation removal, high order
cumulant, timing structure of signal and so on. Compared to the other methods, deviation
removal technology has strong stability and its application range is also wide. FastICA algorithm
based on maximum likelihood estimation is relatively mature, and has good separation
performance and good robustness. Therefore, the FastICA algorithm based on maximum
likelihood deviation removal is proposed to reduce the noise effect in the image by modifying
the deviation[4].

Dept. of ECE, SJCIT Page 9 2019-2020

Night Vision Technology

CHAPTER 4

PERFORMANCE ANALYSIS

Experiments were conducted on the standard 256 * 256, apple images collected during
day and at night. Artificial noise was added to the noise-free images, then apply Median-Average
Filtering, FastICA algorithm based on Maximum Likelihood estimation and the modified
FastICA algorithm to denoise the noisy images. As noise-free images, they are added to the
Gaussian noise that the probability is σ2=0.03 and Salt and Pepper noise that the probability is
P=0.03. The simulation is shown in Fig 4.1

Fig 4.1: The pure picture added with noise[4]

The median filter is suitable for salt and pepper noise denoising, and the mean filter is suitable
for Gaussian noise. Therefore, this experiment uses three kinds of methods, namely the Median-
Mean filtering method, the FastICA algorithm based on Maximum Likelihood estimation, and
the FastICA algorithm based on maximum likelihood deviation removed, to simulate the image
denoising process through Matlab2012b. And simulation test results are shown

Fig 4.2: The median-mean filtering denoising method[4]

Dept. of ECE, SJCIT Page 10 2019-2020

Night Vision Technology

Fig 4.3: The FastICA denoising method[4]

Fig 4.4: The improved fastICA algorithm denoising method[4]

It can be seen from the visual simulation that the noise reduction image simulated by the
modified algorithm is less, and more clear than the other two methods’ results. In order to
objectively evaluate the denoising effect of the 3 methods, the comparisons of the Relative Peak
Signal to-Noise Ratio (RPSNR) and the running time.

In the formula, fij is the original image, ij is the noise reduction image, and the size of image is
M×N .The larger the PSNR is, the better the image’s quality is, which is closer to the original
image.

Dept. of ECE, SJCIT Page 11 2019-2020

Night Vision Technology

CHAPTER 5

ADVANTAGES AND APPLICATIONS

5.1 ADVANTAGES

 The biggest advantage is just having ability to see in dark.

 Its working is simple and easy to use.
 It is designed for longer life.
 Night vision devices are compact in size and easily portable.
 Able to detect people and vehicles at greater distances.

5.2 APPLICATIONS

 Military: In military it is used to keep eyes on unwanted activities and unwanted things at
boarder or at a specific area.
 Law enforcement: It is used by the government officers to look after the details of place
at night where we are unable to see.
 Hunting: In forests it is very difficult to see the things at dark night so by using this
device one can identify the object in dark also.
 Wildlife observation: For taking care of wildlife animals and to keep observation on
illegal activities in wildlife this is used.
 Navigation: Used to show the way and also obstacles in path. This mainly observed in
automobiles and ships.
 Hidden object detection: By using thermal imaging process it is possible to detect the
things buried under earth surface.
 Surveillance: In night vision cameras are mounted around a factory or house to get aware
from surroundings.

Dept. of ECE, SJCIT Page 12 2019-2020

Night Vision Technology

CHAPTER 6
CONCLUSION
The original purpose of night vision was to locate enemy targets at night. It is still used
extensively by the military for that purpose, as well as for navigation, surveillance and targeting.
Police and security often use both thermal-imaging and image-enhancement technology,
particularly for surveillance. The purpose of image denoising is to reduce the impact of the noise
on the images ,there are many denoising algorithms to remove the noise but FastICA and the
modified FastICA are mostly equal, and both of them are higher than the Median-Mean filtering
method in the averaged time-consuming for the reason that the noise reduction process requires
iteration consumption. In general, the modified FastICA based on maximum likelihood deviation
removed has the unique advantage for night vision image denoising.

Dept. of ECE, SJCIT Page 13 2019-2020

Night Vision Technology

REFERENCES
[1] P.R.Namitha and K.Mounika. Study of Night Vision Technology, IJCRST vol.4, Issue
2,2018.
[2] Mohd Junedul Haquel and Mohd Muntjir. Night Vision Technology: An Overview, IJCA
vol.167, pp 37-42,2017.
[3] Rupesh P.Raghatate and et.al Night Vision Techniques and Their Applications, IJMER,vol. 3
Issue 2,pp 816-820,2013.
[4] Hongiun Wang, Duan,Hui Zhao and Youjun Yue. Research of Night Vision Image Denoising
Method Based on the Improved FastICA, IEEE 2017.
[5] Soundarya. P, Dr. Indumathy. R. A Suyvey on Night Vision Techniques and Their
Applications, IJIRCCE Vol.4, Issue8, 2016.
[6] Weikuan Jia,Yuanjie Zheng and Xiang Yin. Preprocessing method of night vision image
application in apple harvesting robot, Int J Agric & Biol Eng vol.11, pp 158-163, 2018.
[7] Jia W K, Zhao D A, Ruan C Z, Liu X Y, Chen Y Y, Ji W. Combination method of night
vision image denoising based on wavelet transform and ICA. Transaction of the CSAM,2015.
[8] Chengzhi Ruan,Chen Chen and Tian shen. Night Vision Image De-noising of Apple
HarvestingRobotsbasedontheWaveletFuzzyThreshold.IntJAdvRobotSyst,2015.
[9] Malarvizhi.D, Lavanya.v, Nivetha priya. M. Night Vision Technology. IJSART Vol 3, Issue8,
2017.
[10] Bhawna Goyal and Ayush Dogra. A Survey on the Image De-noising to Enhance Medical
Images. Biosci., Biotech.Res.Asia,Vol15(3), pp 501-507, 2018.
[11] Verma, Rohit and Jahid Ali. A Comparative study of various types of imagr noise and
efficient noise removal techniques. International Journal of advanced research in computer
science and software engineering, 2013.

Dept. of ECE, SJCIT Page 14 2019-2020