Professional Documents
Culture Documents
59.DETECTING-THE-MOVEMENT-OF-OBJECTS-WITH-WEB-CAM_compressed
59.DETECTING-THE-MOVEMENT-OF-OBJECTS-WITH-WEB-CAM_compressed
2Assistant professor, Department of CSE, Sri Indu Institute of Engineering & Technology, Hyderabad
3,4,5,6 IVth Btech Student, Department of CSE, Sri Indu Institute of Engineering & Technology, Hyderabad
ABSTRACT
Various Methods Are Used In Motion Detection Of A Particular Interest. Each Algorithm Is
Found Efficient In One Way. But There Exists Some Limitation In Each Of Them. This Paper
Proposes A Method For Detecting The Motion in a Particular Region Being Observed. The
Motion Tracking Surveillance Has Gained A Lot Of Interests Over Past Few Years. This System
Is Brought Into Effect Providing Relief To The Normal Video Surveillance System Which
Offers Time-Consuming Reviewing Process. Through The Study And Evaluation Of Products,
We Propose A Motion Tracking System Consisting Of Its Method For Motion Detection . In
Our Proposed System Those Disadvantages Are Omitted And Combining The Usage Of Best
Method We Are Creating A New Motion Detection Algorithm For Our Proposed Motion
Tracking System.
1
pointing to the best prediction macroblock In this we are going to write a python
in a reference frame or field. program which is going to analyse the
images taken from the webcam and try to
EXISTING SYSTEM: detect the movement.Videos can be treated
Digital surveillance systems are mostly as a stack of pictures called frames. Here I
specifically designed for commercial use am comparing different frames(pictures)
and it has always been out of reach for to the first frame which should be static(No
other users. The cost for CCD movements initially). We compare two
cameras,networking devices and the images by comparing the intensity value of
software designed for this system has each pixel.In my project,I used Python
made it inaccessible and impractical for Programming Language and its most
home users with moderate requirements. important and specific libraries OpenCV
Also, not all the existing products have the which is most required for solving
motion detection function. In traditional problems related to images and videos and
systems for security operations, cameras this is an Open Source Computer Vision
are used to deliver analogue video images based personal project to detect Human
to monitors or time-lapse video cassette Faces and different objects coming in front
recorders (VCR). Although many local of the webcam for a specific time
image processing functions are possible to frame.This python scripts detects
improve the system application, this movement on your web-cam and outlines
requires a lot of processing resources and the moving object on your computer
high-powerconsuming hardware. screen.
Although Digital video surveillance and Advantages:
security systems are widely used, analogue
systems still serve as a cheaper alternative. 1. Requires less memory.
2
advancements have improved the motion detection is basically the process of
functioning of motion detectors, the basic comparing sequential images and
purpose and system of this device remains determining whether the differences
the same. A motion detector is basically between them represent motion.
designed to detect motion or movement in
a designated area and may instantly send a 3. SYSTEM REQUIREMENTS:
signal to the security alarm system. A
motion detector is a device that detects 3.1 HARDWARE REQUIREMENTS:
moving objects, particularly people. Such a System : MINIMUM i3.
device is often integrated as a component Hard Disk : 40 GB.
of a system that automatically performs a
task or alerts a user of motion in an Ram : 4 GB.
area.energy efficiency, and other useful 3.2 SOFTWARE REQUIREMENTS:
systems.
Operating System : Windows 8
How motion detection cameras work:
Coding Language : Python 3.7
Monitoring your data centers and building
3.3 SYSTEM STUDY
access points with CCD cameras can
represent a boon to your overall security, FEASIBILITY STUDY
but you need to know a little about camera
terminology to make the best use of them. The feasibility of the project is
Don't think about "aim and shoot" when analyzed in this phase and business
you think about these cameras. Think, proposal is put forth with a very general
instead, about how you might go about plan for the project and some cost
detecting motion in a series of still images. estimates. During system analysis the
First off, let's examine the term "CCD". This feasibility study of the proposed system is
stands for "charge-coupled device". A CCD to be carried out. This is to ensure that the
is a silicon chip with a surface that is proposed system is not a burden to the
divided into light-sensitive pixels. When company. For feasibility analysis, some
light hits these pixels, tiny electric charges understanding of the major requirements
are generated. With enough of these pixels, for the system is essential.
you can get a fairly high resolution image.
Three key considerations involved in the
With adequate "sensitivity" you can detect
feasibility analysis are
motion even in a dimly lit room.
ECONOMICAL FEASIBILITY
The difference between digital cameras
and security cameras is in their basic way TECHNICAL FEASIBILITY
of working. Digital cameras store images
when you tell them to. Security cameras SOCIAL FEASIBILITY
only provide images (save and transmit
them) when they detect motion. And
3
ECONOMICAL FEASIBILITY His level of confidence must be raised so
that he is also able to make some .
This study is carried out to check
the economic impact that the system will
4. SYSTEM DESIGN
have on the organization. The amount of
fund that the company can pour into the 4.1 IMPLEMENTATION: MODULES:
research and development of the system is
limited. The expenditures must be 1.Add Product Details
justified. Thus the developed system as To build project I used some sample
well within the budget and this was products image to train product
achieved because most of the technologies
identification models
used are freely available.
2. Train Model
Only the customized products had to be
purchased. In this Module screen train model
generated with 100% accuracy and now
TECHNICAL FEASIBILITY
show product to web cam.
This study is carried out to check the 3. Add/Remove Product from basket
technical feasibility, that is, the technical
requirements of the system. Any system To allow application to identify product
developed must not have a high demand on image and then show in text area and if we
the available technical resources. This will again show same product then application
lead to high demands on the available will remove from text area
technical resources. This will lead to high
demands being placed on the client. The 5.SOFTWARE ENVIRONMENT
developed system must have a modest
requirement, as only minimal or null What is Python :Below are some
changes are required for implementing
facts about Python.
this system.
Python is currently the most widely used
SOCIAL FEASIBILITY
multi-purpose, high-level programming
The aspect of study is to check the language.
level of acceptance of the system by the
Advantages of Python :-
user. This includes the process of training
the user to use the system efficiently. The Let’s see how Python dominates over other
user must not feel threatened by the languages.
system, instead must accept it as a
necessity. The level of acceptance by the 1. Extensive Libraries
users solely depends on the methods that
Python downloads with an extensive
are employed to educate the user about the
library and it contain code for various
system and to make him familiar with it.
4
purposes like regular expressions, 7. Readable
documentation-generation, unit-testing,
web browsers, threading, databases, CGI, Because it is not such a verbose language,
email, image manipulation, and more. 2. reading Python is much like reading
Extensible English. This is the reason why it is so easy
to learn, understand, and code.
As we have seen earlier, Python can be
extended to other languages. You can write 8. Object-Oriented
some of your code in languages like C++ or This language supports both the
C. This comes in handy, especially in procedural and objectoriented
projects. programming paradigms. While functions
3. Embeddable help us with code reusability, classes and
objects let us model the real world. A class
Complimentary to extensibility, Python is allows the encapsulation of data and
embeddable as well. You can put your functions into one.
Python code in your source code of a
different language, like C++. This lets us 9. Free and Open-Source
add scripting capabilities to our code in the Like we said earlier, Python is freely
other language. available. But not only can you download
4. Improved Productivity Python for free, but you can also download
its source code, make changes to it, and
The language’s simplicity and extensive even distribute it. It downloads with an
libraries render programmers more extensive collection of libraries to help you
productive than languages like Java and with your tasks.
C++ do. Also, the fact that you need to write
less and get more things done. 5. IOT 10. Portable
Opportunities When you code your project in a language
Since Python forms the basis of new like C++, you may need to make some
platforms like Raspberry Pi, it finds the changes to it if you want to run it on
future bright for the Internet Of Things. another platform. But it isn’t the same with
This is a way to connect the language with Python.
the real world. Advantages of Python Over
6. Simple and Easy Other Languages 1. Less Coding
When working with Java, you may have to Almost all of the tasks done in Python
create a class to print ‘Hello World’. But in requires less coding when the same task is
Python, just a print statement will do. It is done in other languages. Python also has
also quite easy to learn, understand, and an awesome standard library support, so
code. you don’t have to search for any thirdparty
libraries to get your job done 2. Affordable
5
Python is free therefore individuals, small 4. Underdeveloped Database Access
companies or big organizations can Layers
leverage the free available resources to
build applications Compared to more widely used
technologies like JDBC (Java DataBase
3. Python is for Everyone Connectivity) and ODBC (Open DataBase
Connectivity)
Python code can run on any machine
whether it is Linux, Mac or Windows. 5. Simple
Programmers need to learn different
languages for different jobs but with No, we’re not kidding. Python’s simplicity
Python, you can professionall can indeed be a problem. Take my example.
I don’t do Java, I’m more of a Python
Disadvantages of Python person. To me, its syntax is so simple that
the verbosity of Java code seems
So far, we’ve seen why Python is a great unnecessary.
choice for your project. But if you choose it,
you should be aware of its consequences as 6. History of Python
well. Let’s now see the downsides of
choosing Python over another language. What do the alphabet and the
programming language Python have in
1. Speed Limitations
common? Right, both start with ABC. If we
We have seen that Python code is executed are talking about ABC in the Python
line by line. But since Python is context, it's clear that the programming
interpreted, it often results in slow language ABC is meant. ABC is a
execution. This, however, isn’t a problem generalpurpose programming language
unless speed is a focal point for the project. and programming environment, which had
been developed in the Netherlands,
2. Weak in Mobile Amsterdam, at the CWI (Centrum
Computing and Browsers Wiskunde &Informatica). The greatest
achievement of ABC was to influence the
While it serves as an excellent server-side
language, Python is much rarely seen on design of Python.Python was
the client-side. Besides that, it is rarely conceptualized in the late 1980s. What is
ever used to implement smartphonebased Machine Learning : -
applications. One such application is called
Carbonnelle. Before we take a look at the details of
various machine learning methods, let's
3. Design Restrictions start by looking at what machine learning
is, and what it isn't. Machine learning is
As you know, Python is dynamicallytyped.
often categorized as a subfield of artificial
This means that you don’t need to declare
intelligence, but I find that categorization
the type of variable while writing the code.
can often be misleading at first brush. The
6
study of machine learning certainly arose to go. The reason behind is that ML has not
from research in this context, but in the been able to overcome number of
data science application of machine challenges. The challenges that ML is
learning methods, it's more helpful to think facing currently are –
of machine learning as a means of building
Applications of Machines Learning
models of data.
:applications of ML − Emotion analysis
Categories Of Machine Leaning :- Sentiment analysis
Supervised learning involves somehow Error detection and prevention
modeling the relationship between
measured features of data and some label Weather forecasting and prediction How
associated with the data; once this model is
to start learning ML?
determined, it can be used to apply labels
to new, unknown data. This is This is a rough roadmap you can follow on
furthersubdivided into classification tasks your way to becoming an insanely talented
and regression tasks: in classification, the Machine Learning Engineer. Of course, you
labels are discrete categories, while in can always modify the steps according to
regression, the labels are continuous your needs to reach your desired end-goal!
quantities. We will see examples of both
types of supervised learning in the Step 1 – Understand the Prerequisites
following section.
In case you are a genius, you could start ML
Need for Machine Learning:- directly but normally, there are some
prerequisites that you need to know which
Human beings, at this moment, are the include Linear Algebra,
most intelligent and advanced species on Multivariate Calculus, Statistics, and
earth because they can think, evaluate and Python. And if you don’t know these, never
solve complex problems. On the other side, fear! You don’t need a Ph.D. degree in these
AI is still in its initial stage and haven’t topics to get started but you do need a
surpassed human intelligence in many basic understanding.
aspects. Then the question is that what is
the need to make machine learn? The most (a) Learn LinearAlgebra and
suitable reason for doing this is, “to make Multivariate Calculus
decisions, based on data, with efficiency
Both Linear Algebra and Multivariate
and scale”.
Calculus are important in Machine
Challenges in Machines Learning :- Learning. However, the extent to which you
need them depends on your role as a data
While Machine Learning is rapidly scientist.
evolving, making significant strides with
cybersecurity and autonomous cars, this
segment of AI as whole still has a long way
7
(b) Learn Statistics you need to make a weather forecast
model. As the amount of data you have
Data plays a huge role in Machine Learning. keeps growing, your algorithms learn to
In fact, around 80% of your time as an ML make more accurate predictions faster.
expert will be spent collecting and cleaning
data. And statistics is a field that handles 4. Handling multi-dimensional and
the collection, analysis, and presentation of multi-variety data
data. So it is no surprise that you need to
learn it!!! (c) Learn Python Machine Learning algorithms are good at
handling data that are multi-dimensional
Some people prefer to skip Linear Algebra,
and multi-variety, and they can do this in
Multivariate Calculus and Statistics and
dynamic or uncertain environments. 5.
learn them as they go along with trial and
error. But the one thing that you absolutely Wide Applications
cannot skip is Python! While there are
You could be an e-tailer or a healthcare
other languages you can use for Machine
provider and make ML work for you.
Learning like R, Scala, etc. Python is
Where it does apply, it holds the capability
currently the most popular language for
to help deliver a much more personal
ML.
experience to customers while also
Advantages of Machine learning :1. targeting the right customers.
8
3. Interpretation of Results used for both research and production at
Google.
Another major challenge is the ability to
accurately interpret results generated by Numpy
the algorithms. You must also carefully Numpy is a general-purpose
choose the algorithms for your purpose. 4. arrayprocessing package. It provides a
High error-susceptibility highperformance multidimensional array
object, and tools for working with these
Machine Learning is autonomous but arrays.
highly susceptible to errors. Suppose you
train an algorithm with data sets small Pandas
enough to not be inclusive. You end up with
Pandas is an open-source Python Library
biased predictions coming from a biased
providing high-performance data
training set. This leads to irrelevant
manipulation and analysis tool using its
advertisements being displayed to
powerful data structures. Python was
customers.
majorly used for data munging and
Python Development Steps : preparation. It had very little contribution
towards data analysis. Pandas solved this
Guido Van Rossum published the first problem.
version of Python code (version 0.9.0) at
alt.sources in February 1991. This release Matplotlib
included already exception handling,
Matplotlib is a Python 2D plotting library
functions, and the core data types of list,
which produces publication quality figures
dict, str and others. It was also object
in a variety of hardcopy formats and
oriented and had a module system. Python
interactive environments across platforms.
version 1.0 was released in January 1994.
The major new features included in this Scikit – learn
release were the functional programming
tools lambda, map, filter and reduce, which Scikit-learn provides a range of supervised
Guido Van Rossum never liked.Six and a and unsupervised learning algorithms via
half years later in October 2000, Python 2.0 a consistent interface in Python. It is
was introduced. licensed under a permissive simplified BSD
license and is distributed under many
Modules Used in Project :Tensorflow Linux distributions, encouraging academic
and commercial use. Python
TensorFlow is a free and opensource
software library for dataflow and Download the Correct version into the
differentiable programming across a range system
of tasks. It is a symbolic math library, and
is also used for machine learning Step 1: Go to the official site to download
applications such as neural networks. It is and install python using Google Chrome or
9
any other web browser. OR Click on the Step 4: Scroll down the page until you find
following the Files option.
link: https://www.python.org
Step 5: Here you see a different version of
Now, check for the latest and the correct python along with the operating system.•
version for your operating system. Step 2: To download Windows 32-bit python, you
can select any one from the three options:
Click on the Download Tab. Windows x86 embeddable zip file,
Windows x86 executable installer or
Windows x86 web-based installer.
10
• To download Windows 64-bit python, you Step 3: Open the Command
can select any one from the three options: prompt option.
Windows x86-64 embeddable zip file,
Windows x86-64 executable installer or Step 4: Let us test whether the python is
Windows x86-64 web-based installer. correctly installed. Type python –V and
press Enter.
Installation of Python
11
Step 3: Click on IDLE (Python 3.7 64-bit)
and launch the program
7. SCREENSHOTS
8. BIOBIOGRAPHY
8. BIOBIOGRAPHY
Although it is implemented, the current
algorithm implemented is much more
efficient than that of before. It just takes the
input frame of images and detects the
difference considering it as the motion
detection.Alert systems may be
implemented automatically when the
motion is detected.Outcome of the project
12
is to detect the motion through a webcam
with much higher accuracy.
previously implemented motion detection restoration, involves the correction of
algorithms.few things concluded distortion, degradation, and noise
fromproject which are absent in many introduced during the imaging process.
previously implemented algorithms are... Interpolation is the Technique mostly used
for tasks such as zooming, rotating,
● Requires less memory shrinking, and for geometric corrections.
● Saves a lot of time Removing the noise is an important step
when processing is being performed. . If
● Most reliable the features extracted are carefully chosen
it is expected that the features set will
● No need of manual monitoring extract the relevant information from the
● Analysis is done automatically input data in order to perform the desired
task using this reduced representation
● Archive space to store videos instead of the full size input. Modules 1.
isn’t needed. Graphical User Interface 2. Dataset Upload
3. Image Processing Module
● Alert systems may be implemented
automatically when the motion is 9.2 Tools and Technologies Used
detected.
OpenCV Library OpenCV (Open Source
Computer Vision Library) is an open
9. IMPLEMENTATION
source computer vision and machine
9.1 Overview of the project learning software library. OpenCV is a free
open source library used in real-time
Image Acquisition Performing image image processing OpenCV is a great tool for
acquisition in image processing is always image processing and performing
the first step in the workflow sequence be- computer vision tasks.It is an opensource
cause, without an image, no processing is library that can be used to perform tasks
possible. After the image has been like face detection, objection tracking,
obtained, various Methods of processing landmark detection, and much more. In
can be applied to the image to perform the this application OpenCV library we are
many different vision tasks. There are using for Feature extraction of the Indian
various ways to acquire image such as with Currency. Feature extraction is a special
the help of camera or scanner. Acquired form of dimensional reduction. When the
Image should retain all the features. Pre- input data to an algorithm is too large to be
Processing The main goal of the pre- processed and it is suspected to be very
processing to enhance the visual redundant then the input data will be
appearance of images and improve the ma- transformed into a reduced representation
nipulation of data sets. Image set of features.
preprocessing, also called image
13
9.3.1 Algorithm # issue is really that Django is missing
"currencydetection.settings")
try:
ImportError:
14
10. TESTING 10.2.3 Functional test
15
10.2.6 Black Box Testing extracted including various domination
parts of note (like identification marks of
Black Box Testing is testing the software the currency).The features are extracted
without any knowledge of the inner using image based segmentation using
workings, structure or language of the template matching and works well in the
module being tested. Black box tests, as whole process with less computation time.
most other kinds of tests, must be written The complete methodology works for 100,
from a definitive source document, such as 500 and 1000 currency notes. The method
specification or requirements document, is very simple and easy to implement. This
such as specification or requirements technique is very adaptive to implement in
document real time world.
10.2.7 Unit Testing The process begins from image acquisition
and end at comparison of features.
Unit testing is usually conducted as part of
a combined code and unit test phase of the 11.1 Future Enhancement
software lifecycle, although it is not In this project, we actually worked on 5
uncommon for coding and unit testing to currencies and it was found that the
be conducted as two distinct phases. 10.3 proposed algorithm based on color and
Test strategy and approach feature analysis works well for four
Field testing will be performed manually currencies. The USD Banknotes has very
and functional tests will be written in minute differences, when it comes to size
detail. and color, making it difficult to recognize
using proposed algorithm. But for INR,
10.3.1 Test objectives AUD, SAR, the algorithm works very well.
So we can conclude that with above
All field entries must work properly. Pages proposed algorithm, we can recognize
must be activated from the identified currencies, which have good
link.The entry screen, messages and differentiation in color and size related to
responses must not be delayed. 10.3.2 features. Also NN based approach give
quicker results. We have also successfully
Features to be tested
retrieved currency conversion rates from
Verify that the entries are of the correct the internet, using web service.
format.No duplicate entries should be
allowed .All links should take the user to 12.REFERENCES
the correct page. 1. Kumar, B. V., Sharan, D. S. (2007,
December). Pattern recognition with
11.CONCLUSION localized gabor wavelet grids. In
In this technique, the authentication of International Conference on
currency is described by applying image Computational Intelligence and
processing. Basically some features are
16
Multimedia Applications (ICCIMA information filtering,ˆa vol. 27, pp.
2007) (Vol. 2, pp. 517-521). 841ˆa842, Aug 2010.
2. IEEE. F. Sebastiani, ˆaMachine learning 6. Guo, J., Zhao, Y., Cai, A. (2010,
in automated text categorization,ˆa vol. September). A reliable method for
34, p. 6, Aug2002.Liu X. Luo M. Zhang paper currency recognition based on
P. Wang W. Huang, W. video based LBP. In 2010 2nd IEEE
abnormal driving behavior detection InternationalConference on Network
via deep learning fusion. In IEEE In- frastructure and Digital Content
Access, 7, 64571-64582., pages 1–12, (pp. 359-363). IEEE. F. Bonchi and E.
2019.
7. Ferrari, ˆaShort text classification in
3. Zhang, J., Wu, L. (2009, November). An twitter to improve information
improved method for image edge filtering,ˆa vol. 27, pp. 841ˆa842, Aug
detection based on GM (1, 1) model. In 2010.Aaqib Saeed, Stojan Trajanovski,
2009 International Conference on Maurice van Keulen, and Jan van Erp.
Artificial Deep erate
17