IRJET- Surveillance of Object Motion Detection and Caution System using Block Matching Algorithm

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 05 Issue: 03 | Mar-2018 www.irjet.net p-ISSN: 2395-0072
© 2018, IRJET | Impact Factor value: 6.171 | ISO 9001:2008 Certified Journal | Page 1474
SURVEILLANCE OF OBJECT MOTION DETECTION AND CAUTION SYSTEM
USING BLOCK MATCHING ALGORITHM
P Jefrin sounder1, K Sudhahar2, N Gowri vidhya3, S Shalini4
1,2UG scholar, Department of Computer Science and Engineering, Prince Dr. K. Vasudevan College of
Engineering and Technology, Chennai, India.
3,4Assistant professor, Department of Computer Science and Engineering, Prince Dr. K. Vasudevan College of
Engineering and Technology, Chennai, India.
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract— A monitoring and alerting application is
introduced in this paper. IP Cameras are special cameras that
stream the video feed into the internet as a video stream. This
stream is captured and can be monitoredforanyunauthorized
activities. Motion detection is done when movements are
detected and the alarm goes ON if such movements are
detected. We can start the alarm feature by entering a
password and clicking on start button. Enabled for both local
cameras as well as IP cameras. Remote monitoring from any
part of the world using a very efficient motion detection
algorithm for better security. The motion detection module
consists of the motion detection algorithm which helps us to
analyze the camera feed and to detect and signal any motion
related triggers. It also comes with a motion sensitivity panel
were you get to adjust the level of motion sensitivity that
might be required. Block matching algorithm is used for
motion detection and frame validation in the system.
Keywords - Motion detection, Block matching algorithm,
frame validation.
I. INTRODUCTION
A new era of computing technology evolving recently
is the Internet of Things(IoT). Machine to machine, machine
to infrastructure, machine to environment, the Internet of
Everything, the Internet of Intelligent Things, intelligent
systems call it what you want, but it’s happening, and its
potential is huge. We see the IoT as billions of smart,
connected “things” that will encompass every aspect of our
lives, and its foundation is the intelligence that embedded
processing provides. The IoT is comprisedofsmartmachines
interacting and communicatingwithothermachines,objects,
environments and infrastructures. As a result, hugevolumes
of data are being generated, and that data is being processed
into useful actions that can “command and control”thingsto
make living much easier and safer and to reduce our impact
on the environment. The creativity of this new era is
boundless, with amazing potential to improve standards of
living.
The Internet of Things(IoT) is thenetworkofphysical
objects devices, vehicles, buildings and other items
embedded with electronics, software, sensors, and network
connectivity that enable these objects to collect and
exchange data. Environmentalmonitoringapplicationsofthe
IoT typically use sensors to assist in environmental
protection by monitoring air or water quality, atmospheric
or soil conditions, and can even include areas like
monitoring the movementsof wildlife andtheirhabitats. The
ability to network embedded devices with limited CPU,
memory and power resources means that IoT finds
applications in nearly every field. Such systems could be in
charge of collecting information in settings ranging from
natural ecosystems to buildings and factories, thereby
finding applications in fields of environmental
sensing and urban planning. Monitoring and controlling
operations of urban and rural infrastructures like bridges,
railway tracks, on- and offshore- wind-farms is a key
application of the IoT.
The problem dealswith the camera functionalitywith
sensitive areas. In this problem the future is to be predicted
earlier and footage should be stored in a data setorastorage
device. The problem case requires 24/7monitoringtodetect
any unusual activities occurring in the footage. The problem
requires large data storage to save every feed from the
camera asit is a 24/7 footage it needsdata storage of atleast
in terabits size. The problem only providesmoderatelevelof
security; hence the system needs more secure operations
and monitoring. In case of sensors, it is very expensive and
implementation of sensors is limited as it operates only
within small range which makes sensors unusable in large
field of monitoring applications.
II.RELATED WORK
[2] This paper proposes a system for the motion
characteristics estimation of multiple objects withuncertain
quantity. The system employs background subtraction
method based on Gaussian Mixture Model to detect objects
and tracks them through an improved algorithm combining
Camshift with Kalman filtering. By analyzing the directions
and trails of targets, the system can figure out their pixel
distance and velocity. Experiment conducted realizes the
proposed system, verifying its performance on motion
characteristics estimation. For the detection of moving
objects, we adopt background subtraction method based on
Gaussian mixture model and introduceself-adaptivelearning
rate to timely update the background models. After
extracting the targets in foreground, Process the binary
image based on morphology operations and connected
domain analysis for further improvement of accuracy in
subsequent procedure.

[3] Moving item recommendation is now and again a
champ among the most key and basic procedures in PC
perspective programs. In this paper, a novel organization
model is suggested to get rid of illustrating closer see things
from files that may contain different sorts of disturbing
impact, for instance, light changes, camera parameter
varieties, bustles and component back-grounds,etc.Foreach
appearance, an nearby do it again reaction guide is created
using temporary double wavelet story distinguish in
community areas, and by expelling the interaction among
communities of the reaction summarize, reduced pixel
emphasize is showed as close-by do it again plan. By then, a
versatile probabilistic evaluation of pixel emphasize
development changed from piece width evaluation is
performed to determine the prospect of a pixel being
organization. Test assessments on complicated moments of
understanding files illustrate that the suggested system has
stored deserving results.
[4] Motion detection is of paramount importance in
video surveillance systems. In this paper, a novel algorithm
is proposed to extract the exact boundaries of moving
objects in video frames. Using the concepts of Cross-
Correlation and Edge Detection, two well-known motion
detection methods are combined to extract the moving
regions more accurately. Also, these two methods are
modified in terms of accuracy and processing time in order
to make them suitable for algorithm. The experimental
results prove that the proposed method significantly
outperforms the other state-of-the-art schemes in terms of
both objective and subjective performance. Another
advantage of the suggested method over the others is its
relatively low computational complexity and high
implementation speed.
[5] A novel full-image guided filtering method is
proposed. Different with many existingneighborhoodfilters,
all input elements are employed during the proposed
filtering approach.
In addition, a novel scheme called weight propagation is
proposed to compute support weights. It fulfills the
requirements of edge preserving and low complexity. It is
applied to the cost-volume filtering in the local stereo
matching framework. The algorithm utilizing the proposed
filtering method is currently one of the best local algorithms
on the Middlebury stereo tested in terms of both speed and
accuracy. Filters with fixed-size convolution kernel, e.g.,
uniform (box filters) or Gaussian, can be used.
[6] New and more natural human-robotinterfacesare
of crucial interest to the evolution of robotics. This paper
addresses continuous and real-time hand gesture spotting,
i.e., gesture segmentation plus gesture recognition. Gesture
patterns are recognized by using artificial neural networks
(ANNs) specifically adapted to the process of controlling an
industrial robot. Since in continuousgesture recognitionthe
communicative gestures appear intermittentlywiththenon-
communicative, we are proposing a new architecture with
two ANNsin seriesto recognize both kindsof gesture.Adata
glove is used as interface technology. Experimental results
demonstrated that the proposed solution presents high
recognition rates(over 99%for a library of ten gestures and
over 96% for a library of thirty gestures), low training and
learning time and a good capacity to generalize from
particular situations.
III.EXISTING SYSTEM
The existing system provide variety of unwanted and
heavy componentswhich make the applicationveryhardfor
usage, and it is also not fit for the actual usage purpose. They
provide limitation for user connectivity throughlicensing of
the developed software. This makes us unable to utilize the
root purpose of the software at ease. Existing systems is
difficult for the users to understand the design and working
of the software and normally it is harder to use an inefficient
methodology. Also local camera capabilityislimitedandonly
remote camera monitoring facility is enabled. Gesture
detection is done by sliding window algorithm. The feed
from the camera is monitored 24/7 by a human. Different
micro controllersare used to achievemotiondetection.More
number of sensors are used to detect the motion.
The existing systemshave a very large stabilityissues
which are very annoying, hence the existing system is
unstable. The local camera isnot supported,thustheexisting
system doesn’t have locality support. The existing systems
do not have a very good efficient motion detection due to
which people don’t want to use them in most cases which
makes it inefficient. Future is to be predicted earlier and
should be stored in a data set. It requires 24/7 monitoringto
detect any unusual activities. It requires a large data storage
to save every feed from the camera. It only provides
moderate level of security. In case of sensors, it is very
expensive and implementation of sensors is limited as it
operates only within small range which makes sensors
unusable in large field of monitoring applications.
IV.PROPOSED SYSTEM
The proposed system is very efficient software that
can be used to detect and monitor securitywitheasewithout
the need of monitoring the system 24/7. The simple
architecture of the proposed system makes it user friendly
for the usage and implementation of the system. The
proposed system does this by using a very efficient motion
detection algorithm named block matching algorithmwitha
variable sensitivity screening namely sensitive and non-
sensitive screening which makes it vital for the security
purpose, there-by increasing the overall performance of the
system. Both the sensitive and non-sensitive screenings are
flexible and can be extended accordingly to the resolution
and the pixel strength of the display and the camera used.
The sensitive region can be set using an flexibleboxavailable
over the current video stream.
The proposed system is very efficient software than
the existing system as the proposed system doesn’t use any

form of sensor to detect the motion occurred in the
monitored region. The sensitive screening can be manually
fitted by the user according to the region to be monitored
where more level of security is needed. The area which
doesn’t come under the sensitive screening is automatically
considered as the non-sensitive region. The caution system
is triggered only when motion is detected in the sensitive
region, any motion in the non sensitive region will not
trigger the caution system. A motion detector isadevicethat
detects moving objects, particularly people. Such a device is
often integrated as a component of a system that
automatically performsa task or alerts a userofmotioninan
area. They form a vital component of security, automated
lighting control, home control, energy efficiency, and other
useful systems. This proposed system can be used as an
intruder alarm in home, offices, restricted areasin shopping
malls, banks, sanctuaries etc.
The proposed system works efficiently in detecting
the movements by using block matching algorithm with the
sensitivity screening technique. The proposed system is a
simple design, thus it has a better understandability. This
system is very simple and efficient system which is very
much robust under heavy usage. The proposed system does
not crash even if under much load, thus the system is much
stable than the existing system. Proposed system doesn’t
need a large storage space to save the video stream, as the
video stream is not saved until motion is detected. The
proposed system has reduced cost by eliminating the usage
of sensors. This system does not need 24/7 monitoring as
any changes occurring in the sensitive screening region
triggers the caution system which in turn sends alert to the
user. In the proposed system realtime motion detection is
done using block matching algorithm and instant alerts are
sent to the user.
V.SYSTEM ARCHITECTURE
Design Engineering deals with the various diagrams
for the implementation of project. Design is the creation of
a plan or convention for the constructionofanobject,system
or measurable human interaction such as in architectural
blueprints, engineering drawings, businessprocesses, circuit
diagrams, and sewing patterns. Design has
different connotations in different fields.Softwaredesignisa
processthrough which the requirementsare translated into
representation of the model.
The above figure represents the architecture of the
proposed system. Initially a camera is to be installed facing
the region that has to be monitored. Now the camera
capturesthe video stream and displays it to theuserthrough
a monitor screen. After the successful installation of the
camera the user is prompted by the system to set the
sensitive region of area that has to be monitored for any
motion in the region. The remaining region is automatically
considered as non sensitive area.
Now, the process to detect any movement in the
sensitive area begins. Every frame from the video is
separated or filtered for comparison. The first frame of the
video is set as the frame with which other frames are to be
compared. Every frame from the video is now compared
with the first frame of the video. During frame comparison
the block matching algorithm is used to detect motion.
The block matching algorithm works by separating a
frame into various numbers of blocks where each block is a
collection of pixels. The collection of pixels of each block
depends on the camera used. This is the first step of block
matching algorithm which is said to be division ofpixels.The
second step is pixel validation in where each pixel in each
block is validated for its maximum and minimum intensity.
The value of maximum and minimum intensity of each pixel
is stored. The maximum and minimum intensityofeachpixel
is determined by using the greyscale and colour imagery.
Now, having the maximum and minimum intensity value of
each pixel every block of the frame iscomparedwiththefirst
frame. If there is any change in the pixel intensity values
above the ranged value then there is some motion in the
sensitive region which in turn triggers the caution system.
Any motion detected in the sensitive region triggers
the caution system and also the video stream is saved to the
hard disk. Caution system is an alarm system which sends
instant alerts to the user about the motion that has been
detected in the sensitive region. The user receives two
various alerts as shown. The user receives an email about
the motion been detected and also a mobile alert is given to
the user.
1) Screening System
Screening system module is where the user is
prompted by the system to drag and fix the sensitive region
that has to be monitored. In the proposed system there are
two types of screening namely sensitive screening and non

sensitive screening. The area of interest that is to be
monitored is said to be the sensitive region and the region
other than sensitive region is said to be the non sensitive
region. Screening system module works only after the
successful installation of camera. The region other than the
region which is fixed as sensitive screen is automatically set
asthe non sensitive screen. The screening systemisdynamic
(i.e) it is changeable using the drag function in the system.
2) Frame computation
The frame based computation in the systemisdonein
this frame computation module. A video stream is a
continuous flow of picture of every second. Each second of
the video stream is a frame. The first step in frame
computation is the separation of frames from the video
stream. After separation the first frame is set as the frame
with which other frames are compared. The second step of
frame computation is the frame comparison. In here, every
frame is compared with the first frame.
3) Pixel validation
Pixel validation module is the next step after frame
computation. The working of block matching algorithm
begins with division of pixels. The division of pixels is
separating a frame into various numbers of blocks where
each block is a collection of pixels. The second step is pixel
validation in where each pixel in each block is validated for
itsmaximum and minimum intensity.Thevalueofmaximum
and minimum intensity of each pixel isstored.Themaximum
and minimum intensity of each pixel is determined by using
the greyscale and colour imagery.
4) Motion detection
Motion detection is the module where the pixel
intensity of every pixel in each block that was validated or
determined is compared with the pixel intensity of the first
frame blocks to detect any unusual changes in the pixel
intensity value. The maximum and minimum intensityvalue
of each pixel every block of the frame is compared with the
first frame. If there is any change in the pixel intensityvalues
above the ranged value then there is some motion in the
sensitive region. Detection of motion also enables video
recording and triggers the caution system.
5) Caution system
Caution system module covers the last part of the
proposed system. This module works on alerting the user
regarding the security threat. The caution system is enabled
after detecting changes in the intensity value of the pixels in
any block. The user is alerted in two ways. One is by sending
an email to the registered mail id and another is by giving
mobile alert to the user by alarm sound. The second method
is implemented to make sure that the user is able to know
about the security threat even when in move.
ALGORITHM
The algorithm which is used in theproposedsystemis
the block matching algorithm. The block matchingalgorithm
worksby separating a frame into variousnumbersof blocks
where each block is a collection of pixels. Beforeseparatinga
frame into various numbers of blocks frame computation is
done. That is frame filtering is carried out and the firstframe
is set as the frame with which the other frames has to be
compared. The comparison of frames is done by comparing
the pixel intensity of each block in every frame. The
collection of pixels of each block depends on the camera
used. The first step of block matching algorithm is the
division of pixels. The second step is pixel validation in
where each pixel in each block is validated for its maximum
and minimum intensity. The value of maximum and
minimum intensity of each pixel is stored.Themaximumand
minimum intensity of each pixel is determined by using the
greyscale and colour imagery.
Now, having the maximum and minimum intensity
value of each pixel every block of the frameiscomparedwith
the first frame. If there is any change in the pixel intensity
values above the ranged value then there is some motion in
the sensitive region which in turn triggers the caution
system and enables video recording.
VI.EXPERIMENTAL RESULTS
C Sharp (C#) is a very powerful language thatisavery
common tool for building the windows tools. Her for our
project we have selected this language since it servers to be
the most effective placement since the most commonly used
Operating systems now a days are windows and this runs in
the .NET framework with very much stability. It is very
simple and also doesn’t need any additional frameworks to
be installed before we work on it such as in Java whichneeds
the Java JDK and JRE to be installed before we work on Java
applications.

VII.CONCLUSION AND FUTURE ENHANCEMENT
The application of Remote camera monitoring is a
new area of development, hence the novelty in the current
system is the utilization of the local camera also along with
the remote IP camera that makesthe system ubiquitous.The
application APIs have been designed in such a waythatitcan
be used as interface between wide range of people and
clients. The aspect can be used as a generic platform for
many other security monitoring applications. After the
testing work, advantages of the softwareweredescribedand
suggestions for further enhancementandimprovementwere
discussed.
VIII. REFERENCES
1. “Unsupervised Gesture Segmentation by Motion
Detection of a Real-Time Data Stream”, Miguel A.
Simao, Pedro Neto, and Olivier Gibaru.
2. “Motion Characteristics Estimation of Animals in
Video Surveillance”, Wenhuan Huang, QiliZeng,and
Ming Chen.
3. “Twin wavelet Frame Difference based multiple
Moving Object Detection and Tracking in Video
Surveillance Applications”, Hemlata Soni, and Priya
Gupta.
4. “An Algorithm for Detecting the Exact Regions of
Moving Objects in Video Frames”, Salar Fattahi ,
Masoumeh Azghani, and Farokh Marvasti.
5. “Full-Image Guided Filtering for Fast Stereo
Matching”, Qingqing Yang, Dongxiao Li, Member,
IEEE, Lianghao Wang, and Ming Zhang.
6. “Real-Time and ContinuousHand Gesture Spotting:
an Approach Based on Artificial Neural Networks”,
Pedro Neto, Dário Pereira, J. Norberto Pires,
Member, IEEE and A. Paulo Moreira, Member, IEEE.
7. “Selecting Key Poses on Manifold for Pairwise
Action Recognition”, Xianbin Cao, Senior Member,
IEEE, Bo Ning, Pingkun Yan, Senior Member, IEEE,
and Xuelong Li, Senior Member, IEEE.
8. “Continuous Body and Hand Gesture Recognition
for Natural Human-Computer Interaction”, yale
song, david demirdjian, and randall davis.
9. “Handling Movement Epenthesis and Hand
Segmentation Ambiguities in Continuous Sign
Language Recognition Using Nested Dynamic
Programming”, Ruiduo Yang, Sudeep Sarkar,Senior
Member, IEEE, and Barbara Loeding.
10. “Weakly Supervised Training of a Sign Language
Recognition System Using Multiple Instance
Learning Density Matrices”, Daniel Kelly, John Mc
Donald, Member, IEEE, and Charles Mar

IRJET- Surveillance of Object Motion Detection and Caution System using Block Matching Algorithm

Recommended

More Related Content

What's hot (17)

Similar to IRJET- Surveillance of Object Motion Detection and Caution System using Block Matching Algorithm (20)

More from IRJET Journal (20)

Recently uploaded (20)

IRJET- Surveillance of Object Motion Detection and Caution System using Block Matching Algorithm