In today's competitive environment, the security concerns have grown tremendously. In the modern world, possession is known to be 9/10'ths of the law. Hence, it is imperative for one to be able to safeguard one's property from worldly harms such as thefts, destruction of property, people with malicious intent etc. Due to the advent of technology in the modern world, the methodologies used by thieves and robbers for stealing has been improving exponentially. Therefore, it is necessary for the surveillance techniques to also improve with the changing world. With the improvement in mass media and various forms of communication, it is now possible to monitor and control the environment to the advantage of the owners of the property

1. Motion detection system
ABSTRACT
In today's competitive environment, the security concerns have grown tremendously. In the
modern world, possession is known to be 9/10'ths of the law. Hence, it is imperative for one to
be able to safeguard one's property from worldly harms such as thefts, destruction of property,
people with malicious intent etc. Due to the advent of technology in the modern world, the
methodologies used by thieves and robbers for stealing has been improving exponentially.
Therefore, it is necessary for the surveillance techniques to also improve with the changing
world. With the improvement in mass media and various forms of communication, it is now
possible to monitor and control the environment to the advantage of the owners of the
property. The latest technologies used in the fight against thefts and destruction are the video
surveillance and monitoring. By using the technologies, it is possible to monitor and capture
every inch and second of the area in interest. However, so far the technologies used are passive
in nature, i.e., the monitoring systems only help in detecting the crime but do not actively
participate in stopping or curbing the crime while it takes place. Therefore, we have developed
a methodology to detect the motion in a video stream environment and this is an idea to
ensure that the monitoring systems not only actively participate in stopping the crime, but do
so while the crime is taking place. Hence, a system is used to detect any motion in a live
streaming video and once motion has been detected in the live stream, the software will
activate a warning system and capture the live streaming video.
Introduction
In recent years, motion detection has attracted a great interest from computer vision
researchers due to its promising applications in many areas, such as video surveillance, traffic
monitoring or sign language recognition. However, it is still in its early developmental stage and
needs to improve its robustness when applied in a complex environment. Several techniques
for moving object detection have been proposed among them the three representative
approaches are temporal differencing, background subtraction and optical flow. Temporal
differencing based on frame difference, attempts to detect moving regions by making use of
the difference of consecutive frames (two or three) in a video sequence. This method is highly
adaptive to dynamic environments, but generally does a poor job of extracting the complete
shapes of certain types of moving objects. Background subtraction is the most commonly used
approach in presence of still cameras. The principle of this method is to use a model of the
background and compare the current image with a reference. In this way the foreground
objects present in the scene are detected.

2. The method of statistical model based on the background subtraction is flexible and fast, but
the background scene and the camera are required to be stationary when this method is
applied. Optical flow is an approximation of the local image motion and specifies how much
each image pixel moves between adjacent images. It can achieve success of motion detection in
the presence of camera motion or background changing. According to the smoothness
constraint, the corresponding points in the two successive frames should not move more than a
few pixels. For an uncertain environment, this means that the camera motion or background
changing should be relatively small. The method based on optical flow is complex, but it can
detect the motion accurately even without knowing the background.
Architecture of Project
Description
Motion Detection System is to detect the motion in front of a camera. A Web camera with an
USB interface is used to take images. Images acquired from the web camera are passed to
Personal Computer (PC) via USB port. PC consists of inbuilt USB controllers which will control
externally connected hardware devices. This USB port is interfaced to CPU through PIC bus.
Images acquired by Web camera are processed by PC with the help of MATLAB workspace.
MATLAB process those images with the help of Image Acquisition Toolbox.
After the complete processing of images in MATLAB the result will be displayed on it. Here
Monitor is used as a display device in this project.
Working Principle
For obtain this objective a Web camera is used to capture the video. To process this frames
MATLAB workspace is used. Frames taken by camera are loaded onto MATLAB workspace. Then
those frames are displayed on axis Window of Graphical User Interface (GUI). Now these
images passed through a Manual window to find area of interest. Image taken by camera will
be in 3D format as it is in RGB form. It is difficult to process a RGB images, to reduce processing
complexity of images this frames are converted to gray scale Images.
After conversion of images, they are heightened to improve the brightness and contrast (i.e. to
reduce the noises produced by lighting). Before processing has started a boundary is created
and this boundary can be enlarged. Now this gray scale image is converted into binary format in
order to process further. Now using predefined functions in MATLAB workspace difference
between two frames is calculated. And the resultant images obtained by the difference from
those two frames are used to find the motion detection. When there is any motion the

3. resultant image will have a white portion in the motion area and remaining area will be blank. If
there is no motion then image will be totally blank.
Now the calculations are made to find the motion detected in the place using this calculations.
The area were white portion is calculated by summing 1's in the overall image. Since the 1's will
be obtained at white area in the image. Now all 1's are summed up to find whether the sum
crosses the predefined threshold. If the total exceeds the predefined threshold then a Motion
Detected massage displayed else a No motion Detection message will be displayed
Image Processing Techniques Used:
 Image Conversion
 Image Enhancement
 Image Windowing
 Image Arithmetic
With the Help of Image conversion technique RGB frames are converted into gray scale, for
reducing the complexity of 3D image format. And later converted to Binary format. After
conversion of images, they are heightened to improve the brightness and contrast (i.e. to
reduce the noises produced by lighting) by using Enhancement technique. This technique is
used to display the image on the windows. The perfect difference of two frames can be
calculated with help of image arithmetic technique and then using which the motion of the
object can be detected.
Introduction About Video Surveillance
In 1965, Reporters from different countries suggested to use video as surveillance cameras.
Since the video is cheapest source which can be used in public areas. And video surveillance
started by closed circuit television monitoring (CCTV). When videocassette was released into
market, video surveillance became most popular. Using analog technologies the videos which
are recorded were used as evidence. A analog video surveillance uses a camera, monitor and a
VCR.
Old tube cameras used in analog video surveillance were only useful in day time, and VCR got a
capacity of storing a footage of eight hours only. Those were the disadvantages with the analog
video surveillance, for this owners/employees of a system are supposed to change the video
tape on regular interval. As video tapes are re-used, they don't lost for long period of time. The
other major issue was using it in night times (or) in cloudy days. Due to in sufficient lighting

4. provided to old tube camera they faced some problems. Even thought it was a good idea, and
this technology is not yet vanished till today.
Later a Charged Coupled Device camera (CCD), which had an inbuilt microchip computer
technology. In 1990's video surveillance made a significant progress in technology by
introducing digital multiplexing. As cost of the digital multiplexer is reduced, there was a
radically change in surveillance industries by starting recording large no of cameras at once.
Advantages of digital video recorders are listed below:
 The compression capability of video taping is changing drastically using different video
formats, this allows to store more information on to hard drive
 The cost of a hard drive, which has dropped dramatically in recent years.
 The storage capacity of a hard drive, which has increased dramatically in recent years.
Importance of Digital video surveillance raised as cost of it dropped with the computer
revolution. Using Digital video surveillance, regular changing of video tapes has vanished. Now
footage of month's can be stored into hard disk without any problem. As the video recorded is
in digital form it is far clearer then the grainy images (Image taken by analog technology
cameras). There are different image processing technique for digitally stored images which
enhance it in various ways (add light effects, change colors, reverse black and white) to find the
important features of image.
Requirement of video surveillance
While it is important to understand the various places video surveillance can be used it is also
important to asses the risks involved in the protection of a certain item. In the recent years, as
more and more items such as art are gaining importance, the prices of such things are also
going through the roof. Therefore, technology has come in the forefront for protection and
surveillance of such goods and items.
The following are the statistics of thefts in places such as shops, residences and in public places
in our country, India in a particular year. Of the 50 reported thefts in one year, the breakage of
thefts can be shown as the following:
 Display Cases 19
 Open displays 10
 Pictures 04
 Other displays 02

5.  At night 06
 From stores 02
 Long timescale 04
 Other 03
Even though technology as improved from past few decades, it has still long way to go. From
the above statistics we can clearly say that technology is gaining focus on security surveillance.
By enhancing security surveillance technology many crimes can we stopped.
Motion Detection in Live VIDEO STREAM
When all is said and done, surveillance systems should be a reflection of the real world we live
in. As people become more and more security savvy, they will demand real protection for their
property. The new digital video systems will have to raise that security to a new level. They
should make the customers feel good. Scare off a few troublemakers. And those who do try to
beat the system should face a far greater risk of getting caught. Hence, the new digital video
surveillance systems should be able to provide a high sense of security. The peace of mind can
only be achieved when the person is assured that he will be informed of any thefts of his
property while they are in progress. He would also feel more secure if he can be guaranteed
that the surveillance system that he uses will not only give him evidence against the
perpetrators but also try to stop the thefts from taking place in the first place. Therefore, to
achieve such kind of security Motion Detection in the live video stream is implemented. The
motion detection systems will not only be monitoring the areas of interest but will also keep an
active lookout for any motion being produced.
WORK SPECIFICATION:
Aim: In the project, I aimed to build such a surveillance system, which can not only detect
motion, but will
1. Warn the user of the intrusion and
2. Record the footage of the video from the moment the motion was detected.
Coding Language:
To fulfill aim, I have used strong computing software called MATLAB. Advantage of MATLAB:
Basically the advantage of using MATLAB is that MATLAB is an interpreted language for
numerical computation. It allows one to perform numerical calculations, and visualize the
results without the need for complicated and time consuming programming. MATLAB allows its

6. users to accurately solve problems, produce graphics easily and produce code efficiently.
Disadvantage of MATLAB: The only problem with MATLAB is that since MATLAB is an
interpreted language, it can be slow, and poor programming practices can make it unacceptably
slow. If the processing power of the computing machine is low the MATLAB software takes time
to load and execute any code making the code execute very slowly.
Reason for Selection: MATLAB provides Image Acquisition and Image Processing Toolboxes
which facilitate us in creating a good GUI and an excellent code.
STUDY AND ANALYSIS:
The objective of this work was to develop a surveillance system which would detect motion in a
live video feed and if motion is detected, then to activate a warning system and store the video
feed for future reference and processing purposes. The activation of an alarm would help in
nullifying a threat of security and storing of video provides a proof of such malicious activity.
Keeping the work objective in mind, we firstly developed basic system architecture as shown in
the Figure 2.2.
The system architecture, which we developed, describes how the system component interacts
and work together to achieve the overall system goals. It describes the system operation, what
each component of the system does and what information is exchange. The architecture was
designed for basically getting an idea of how the actual system works and operates.
System Architecture Functioning
The system architecture is going to function in following way:
Capturing the live video feed through a web cam:
To detect motion we first have to capture live video frames of the area to be monitored and
kept under surveillance this is done by using a web cam which continuously provides a
sequence of video frames in a particular speed of FPS (frames per second).
Comparing the current frames captured with previous frames to detect motion:
For checking whether any motion is present in the live video feed, we compare the live video
frames being provided by the web cam with each other so that we can detect changes in these
frames and hence predict the occurrence of some motion.

7. Storing the frames on the memory if motion is detected:
If motion is being detected, we would require storing such motion so that the user can view it
in the near future. This also helps the user in providing a legal proof of some inappropriate
activity since a video coverage can be used as a proof in the court of law.
Indicating through an alarm when the motion is detected:
The user may want to be notified immediately that there has been some intrusion detected by
the software, hence an alarm system is included in the software. This alarm system
immediately activates a WAV file format audio alarm signal if any kind of motion is detected
hence. This helps in preventing any kind of breach of security at that moment of time.
SELECTION CRITERIA OF THE TASKS:
Our work is motion based change detection in .avi video format. Before beginning with the
work, one of the important tasks was deciding the various tasks required to implement the
work. Therefore, we performed a brain storming session and decided various important tasks
which would be required in completion of the work such as:
 Analysis and study of the problem definition,
 Deciding the requirements of the system being developed,
 System architecture containing the following sub function:
 Capturing,
 Comparing
 Storing and
 Indication of motion
 Developing the code and
 Documentation.
After deciding the various important tasks in our work, we decided that the platform on which
we are going to develop our code will be MATLAB. We choose MATLAB because various video
acquisition and analysis functions are pre-defined in MATLAB that would make the
development of our work much easier. Finally, just before we started developing the code, we
designed a rough GUI and created a design, which would suit our needs and perform all
activities, which were desired by us and would be easier to use by anybody.

8. MOTION DETECTION:
Rationale:
The detection of motion essentially requires the user to perform two major steps. They are:
foremost step is to setup the hardware for acquiring the video data in which the motion is to be
detected and the later step is to actually device an algorithm by which the motion will be
detected. The AVI video format is actually an interleave of Audio and Video. The video stream is
stored or acquired as a series of frames occurring in an ordered sequence one after the other.
Acquisition Setup:
The MATLAB programming language is used to store data in the form of matrices. Therefore
MATLAB can provide quick interface with data matrices. The software provides for frame
acquisition from hardware devices such as web cams or digital cameras as long as the devices
are correctly initialized by the programmer. Therefore, in order to allow quick setup with the
image acquisition devices, MATLAB Function directory provides a host of predefined functions
by which the user can inquire about the various different devices currently connected and then
setup the required device with MATLAB so that it can acquire and store data at run time.
IMAGE ACQUISITION TOOLBOX IN MATLAB:
The Image Acquisition Toolbox is a collection of functions that extend the capability of the
MATLAB numeric computing environment. The toolbox supports a wide range of image
acquisition operations, including
1. Acquiring images through many types of image acquisition devices, from professional
grade frame grabbers to USB-based Webcams.
2. Viewing a preview of the live video stream.
3. Triggering acquisitions (includes external hardware triggers).
4. Configuring callback functions that execute when certain events occur.
5. Bringing the image data into the MATLAB workspace.
Many of the toolbox functions are MATLAB M-files. You can view the MATLAB code for these
functions using the statement
type function_name
You can extend the capabilities of the Image Acquisition Toolbox by writing your own M-files, or
by using the toolbox in combination with other toolboxes, such as the Image Processing

9. Toolbox and the Data Acquisition Toolbox. The Image Acquisition Toolbox also includes a
Simulink block, called the Video Input block, that can be used to bring live video data into a
model.
Basic Image Acquisition Procedure
The basic steps required to create an image acquisition application by implementing a simple
motion detection application. The application detects movement in a scene by performing a
pixel-to-pixel comparison in pairs of incoming image frames. If nothing moves in the scene,
pixel values remain the same in each frame. When something moves in the image, the
application displays the pixels that have changed values. To use the Image Acquisition Toolbox
to acquire image data, you must perform the following basic steps:
CHAPTER NO.3
CREATING GRAPHICAL USER INTERFACE (GUI)
GRAPHICAL USER INTERFACE:
GUIDE, the MATLAB Graphical User Interface Development Environment, provides a set of tools
for creating graphical user interfaces (GUIs). These tools simplify the process of laying out and
programming GUIs.
Laying Out a GUI:
The GUIDE Layout Editor enables you to populate a GUI by clicking and dragging GUI
components -- such as buttons, text fields, sliders, axes, and so on -- into the layout area. It also
enables you to create menus and context menus for the GUI. Other tools, which are accessible
from the Layout Editor, enable you to size the GUI, modify component look and feel, align
components, set tab order, view a hierarchical list of the component objects, and set GUI
options. The following topic, Laying Out a Simple GUI, uses some of these tools to show you the
basics of laying out a GUI. GUIDE Tools Summary describes the tools.
Running the Simulation from the GUI:
The GUI Simulate and store results button callback runs the model simulation and stores the
results in the handles structure. Storing data in the handles structure simplifies the process of
passing data to other sub function since this structure can be passed as an argument.

10. When a user clicks on the Simulate and store results button, the callback executes the following
steps: Calls sim, which runs the simulation and returns the data that is used for plotting. Creates
a structure to save the results of the simulation, the current values of the simulation
parameters set by the GUI, and the run name and number. Stores the structure in the handles
structure. Updates the list box String to list the most recent run.
MOTION DETECTION ALGORITHM:
The MATLAB interface allows the user to define the commands to be performed at the run
time. Once the user setup of the video source is complete the algorithm comes into play. The
algorithm is built to take advantage of the strength of MATLAB i.e. to store data as a form of
matrices. The frames acquired are stored in the MATLAB directory as matrix in which each
element of the matrix contains information about the pixel value of the image at a particular
location. Therefore, the pixel values are stored in the workspace as a grid where every element
of the matrix corresponds to an individual pixel value.
Since MATLAB considers each matrix as one large collection of values instead of a bunch of
individual values it is significantly quicker in analyzing and processing the image data. The
algorithm hence checks each frame being acquired by the device with the previously acquired
frame and checks for the difference between the total values of each frame. A threshold level is
set by the user with which the difference of values is compared. If the difference exceeds the
threshold value the motion is said to be detected in the video stream.
Function Explanation:
Function Name:
The Function name Mstart is executed as the Monitor button in the GUI is pressed by the user.
It takes the value of the GUI from the user and updates it in the workspace.
Set:
The set command is used to change the looks and controls available to the user in the GUI. It is
used to change the value of the buttons and is also used to prevent the user from pressing
buttons which cannot logically occur again. As the number of buttons that can be pressed by
the user reduces, the amount of confusion in the users mind will also reduce as the process will
be self guiding thereby reducing the number of errors or bugs and to ensure that user's
experience is hassle free.

11. Video Input:
The video input command is used to setup the video source for the rest of the program to be
run. The code presentation for above snippets of the processes is performed by MATLAB when
the video input command is called into play. The initial four commands are given to inquire
about the presence and status of the camera that has been stated in the program. The
subsequent function is carried out if MATLAB is able to connect and initialize the specified
device. If the connection is successfully established MATLAB just sets its input location to the
device, which has been initialized, and then keeps on catching the frames as input data from
the device.
Uiputfile:
The Uiputfile function is used to allow the user to define the name and storage space of the
output file of the video. This function is essential for two major reasons:
1. It allows the MATLAB to save the file exactly where the user specifies thereby ensuring
the user can easily find the storage location.
2. It allows the user to name the file thereby allowing him to keep a record of each and
every file without the chances of any previous record being overwritten.
Aviobj:
The AVI object command is used to create an object file of the type AVI. The AVI file is a
standard video format with a predefined method of encryption. Therefore, a class file is already
present in MATLAB and the AVI object file defines an instance to create the AVI file in which the
motion is being stored. The object created is set to the filename specified by the user in the
previous function.
Get:
The get function is used to interface with the GUI file. It checks the status and returns the
current value of the GUI button as specified.
Start Vid:
The Start function is used to start the video acquisition device to get the frame from the device
object.

12. Stop Vid:
The Stop function is very important in the video acquisition device. The Start function begins
the video stream entering as input to MATLAB. The stop function will stop this input. If the
function is not used, the video stream continues in the path already started by MATLAB. If the
user will try to use it again, MATLAB will not be able to start it again as the path will be busy. It
will therefore stop the reusability of the program unless the whole of MATLAB reinitializes.
Delete Vid:
The delete function is used to delete the temporary frames stored by MATLAB in the object file.
This function will free up the workspace as well as enable the function to reuse the pathname.
Imaqreset:
The imaqreset function is very important as it resets the acquisition device altogether. It
ensures that the frame buffer in the object is free and completely new at the time the device
restarts. It also resets the device therefore ensuring that there is no device present for
acquisition and the device can be used for other uses.
VIDEO, AUDIO, HELP AND GUI:
This section describes about the further development of the video and the audio units along
with the help and the graphical user interfaces.
VIDEO:
The software produces an AVI video file as it monitors the area in view. Irrespective of the fact
that most modern operating systems would provide various different software's to play video
files in AVI format, the user should be able to view the file without having to switch programs
and searching for it. Hence, it is of the utmost importance that there should be a video player
that plays the video stream that has been produced.
Uigetfile:
The uigetfile function is used to retrieve a file from the hard drive. The function is used here to
ensure that the user has the option : choose the movie file that the user wants to playback.
Although it may seem illogical since only one file is created in each instance, the user may want
to keep a track of other files in the record or may want to play a previously recorded video.

13. Mplay:
The mplay function is a built in video player in the MATLAB library function. It can play a host of
multimedia files as well as be called from the MATLAB command prompt. It has a Graphical
User Interface to play the video file thereby allowing the user to stop, play as well as forward
and rewind the file. The mplay function is predefined and encrypted in MATLAB to be run with
the various programs created by the user. The source code for the player is protected.
AUDIO:
In order for the software to act as a surveillance system it is important to provide a mechanism
to raise an alarm in case motion is detected in the video stream. However, conversely, stealth
may be required in a few cases where alarms may prove more harmful. Therefore, an alarm
function is required which will allow the user to choose the audio function as per his requisites.
Get:
The get function is used to interface with the GUI file. It checks the status and returns the
current value of the GUI button as specified.
Global:
The global function is used to specify that the global variable is being called into play. MATLAB
ensures that every variable is local only to its local function to reduce complexity and to reduce
conflicts between similarly named variables. The global function makes the variable accessible
to all functions in the program to allow different functions to change variables according to the
necessity [11].
Wave read:
The wav read function is a MATLAB function to read and store audio files in the wave audio
format. It can search and find the first RIFF chunk of data in the file. It then opens the file for
the wave player and searches for the next subsequent chunks of data. It does not open the
subsequent chunks but just reads the type of chunk that is present and forwards it to the
player.

14. Audio player:
The audio player function is an audio playing file, which can play a host of audio signals in
MATLAB. It essentially initializes the audio file sent to it by the waveread or auread function
and creates an audio signal object which returns the number of bits each signal takes up.
Play:
The play function starts and runs the audio signal object created in the audio player function. It
plays each audio sample as provided in the audio file as it enters in the audio player. The
following are snippets of the play function of the audio player. It checks for any present errors
and if not, plays the audio sample.
HELP:
The help function is a prerequisite for any good software to ensure that the user can use each
and every function of the program. The help file of any software should be detailed with
examples or instructions about using the software to improve user's interaction with the
software.
Mdhelp:
The mdhelp is a completely new gui file which acts as a popup when called as a function. It
contains a static text box that acts as a frame in which there are instructions stored about how
to use the given software. The text frame contains a step-by-step guide to running and
interfacing with the software to help the user make various decisions about the options he
wants exercise.
GUI:
The modern operating systems allow for almost every program to run using visual icons and
Interfaces. Hence, most users would be put off from using software's that are completely text
based to run. MATLAB provides the programmer with MATLAB GUIDE which is a tool for
generating user interfaces for the programs.
MOTION DETECTION METHOD FOR REAL-TIME SURVEILLANCE
An integration of Optical Flow Method, Temporal Differencing Method and Double Background
Filtering (DBF) Method with Morphological Processing is represented. The main goal of this
algorithm is to separate the background interference and foreground information effectively

15. and detect the moving object accurately. Firstly, temporal differencing method is used to detect
the coarse motion object area for the optical flow calculation. Secondly, the DBF method is
used to get and maintain a stable background image to address changes in environmental
conditions and is also used to reduce the background interference, and locate the moving
object position. Morphological processing is also used and combined with DBF to gain the
better results. Different from the paper, a new improved strategy is proposed which not only
improves the capability of detecting the object in motion, but also reduces the computation
demands.
Real-time detection is important in video surveillance to find the moving objects. The algorithm
used in this project integrates the Temporal Differencing Method, Double Background Filtering
(DBF) method, Morphological Processing Method and Optical Flow Methods to get accurate
results.
Temporal differencing method is used to detect initial common motion areas through which
calculation can be achieved easily and accurately. This method also helps in getting real-time
accuracy.
The DBF method keeps a stable background frame to find the environmental changes. And also
helps to eliminate the background interferences by separating the moving objects from it.
The morphological processing methods are adopted and combined with the DBF to get
improved results. The most attractive advantage of this algorithm is that the algorithm does not
need to learn the background model from hundreds of images and can handle quick image
variations without prior knowledge about the object size and shape. The algorithm has high
capability of anti-interference and preserves high accurate rate detection at the same time. It
also demands less computation time than other methods for the real-time surveillance. The
effectiveness of the proposed algorithm for motion detection is demonstrated in a simulation
environment and the evaluation results are reported here.
OVERVIEW OF THE METHOD:
The method is depicted in the flow chart of Figure 4.1. As can be seen, the whole algorithm is
comprised of four steps:
1. Temporal differencing method, which is used to detect the initial coarse object motion
area;
2. Optical flow detection, which is based on the result of (1) to calculate optical flow for
each frame;

16. 3. Double background filtering method with morphological processing, which is used to
eliminate the background interference and keep the foreground moving information;
4. Motion area detection, which is used to detect the moving object and give the alarming
in time.
The final processing result is a binary image in which the background area and moving object
area are shown as white color, the other areas are shown in black color and the top right corner
is the alarm symbol.
TEMPORAL DIFFERENCING DETECTION METHOD:
Temporal differencing is based on frame difference which attempts to detect moving regions by
making use of the difference of consecutive frames (two or three) in a video sequence. This
method is highly adaptive to static environment. So temporal differencing is good at providing
initial coarse motion areas. The two subsequent 256 level gray images at time t and t+1, I (x, y,
t) and I (x, y, t +1) , are selected and the difference between images is calculated by setting the
adaptive threshold to get the region of changes. The adaptive threshold Td can be derived from
image statistics. In order to detect cases of slow motion or temporally stopped objects, a
weighted coefficient with a fixed weight for the new observation is used to compute the
temporal difference image Id (x, y, t) d as shown in following equations:
Where w is a real number between 0 and 1 which describes the temporal range for difference
images. Ia (x, y, t -1) is initialized to an empty image. In our method, we set d T as the three
times of mean value of Ia (x, y, t +1) and w = 0.5 for all the results. Fig.2 below shows the results
of temporal differencing method under a simulation environment which has a static
background of our laboratory. From the results, we can see that the temporal difference is a
simple method for detecting moving objects in a static environment and the adaptive threshold
d T can restrain the noise very well. But if the background is not static, the temporal difference
method will very sensitive to any movement and is difficult to differentiate the true and false
movement. So the temporal difference method can only be used to detect the possible object
moving area which is for the optical flow calculation to detect real object movement.
OPTICAL FLOW DETECTION METHOD:
Optical flow is a concept which is close to the motion of objects within a visual representation.
The term optical flow denotes a vector field defined across the image plane. Optical flow
calculation is a two-frame differential method for motion estimation. Such methods try to
calculate the motion between two image frames which are taken at interval t at every pixel
position. Estimating the optical flow is useful in pattern recognition, computer vision, and other

17. image processing applications. In this chapter, a optical flow method entitled Lucas-Kanade
Method is introduced.
Lucas-Kanade Method
To extract a 2D motion field, Lucas-Kanade method is often employed to compute optical flow
because of its accuracy and efficiency. Barron compared the accuracy of different optical flow
techniques on both real and synthetic image sequences, it is found that the most reliable one
was the first-order, local differential method of Lucas and Kanade. Liu studied the accuracy and
the efficiency trade-offs in different optical flow algorithms. The study has been focused on the
motion algorithm implementation in real world tasks. Their results showed that Lucas Kanade
method is pretty fast. Galvin evaluated eight optical flow algorithms. The Lucas-Kanade method
consistently produces accurate depth maps, and has a low computational cost, and good noise
tolerance. The Lucas-Kanade method is trying to calculate the motion between two image
frames which are taken at time t and t + dt for every pixel position. As a pixel at location (x,y,t)
with intensity Ia (x,y,t) will have moved by dx, dy and dt between the two frames, the following
image constraint equation can be given:
Simplified Calculation:
The theoretical calculation procedure of the optical flow method is explained in the above
subsection, but for the requirement of practical application, some operation characteristics
between matrices can be used to simplify the complexity of calculation. For the calculation of
invertible matrix in (16), the companion matrix method can be used:
Gradient Operator:
From the operation expression of optical flow, the estimation of the gradient for x-direction, y-
direction and t-direction, has a great influence on the final results of optical flow calculation.
The most common gradient operators used in optical flow calculation are Horn, Robert, Sobel,
Prewitt, Barron and so on. In this paper, a better 3D Sobel operator is used which was proposed
in [26]. This operator uses three different templates to do the convolution calculation for three
frames in a row along the directions of x, y and t and to calculate the gradient along three
directions for central pixels of the template in the middle frame. Fig.3 shows the operators.
Results of Optical Flow Detection:
The optical flow information for every frame of an image is calculated. As shown in Fig.4, the
optical flow of frames It, It+1, , , ,It+n in a period time [t, t+n] are represented as n F1 ,F2 , ,Fn

18. (here i represent i sampling period in t+i expression). The result of optical flow is shown as a
binary image and the adaptive threshold is selected to distinguish the moving pixel from the
still pixel. The pixels whose optical flow values are greater than threshold will be considered as
moving pixels and are shown in white.
Where Fn (i, j) n is the optical flow value, Fn D(i, j) n is the result of optical flow detection and
the adaptive threshold T is select as median value of Fn (i, j) n whose value is above 0. Figure
4.5 shows the results of optical flow which is calculated based on the result of temporal
difference. The simulation environment this time is not a static background which the column
bar curtain is moving caused by the winds. From the results, we can see that the optical flow
with adaptive threshold based on temporal difference reserves the information of moving
object very well. However, because of the background interference of the image, the real
object movement still can not be separated from the background. So the method of double
background filtering with morphological processing is introduced in the next section to deal
with this problem.
DOUBLE BACKGROUND FILTERING WITH MORPHOLOGICAL PROCESSING:
By using the optical flow method, two types of optical flow information are obtained, which are
the interference information of image background and the information of image pixel with any
possibility of real object movement. In the real situation, because of the environment such as
light, vibration and so on, the interference information of the background still can be detected.
Sometimes, it is difficult for the real object movement to be differentiated from the background
interference. In this section, the method of DBF with morphological processing is used to get rid
of the background interference and separate the moving object from it. Firstly, the DBF method
and its corresponding results are discussed. Then the morphological processing methods are
introduced and the improved results are also demonstrated.
Double Background Filtering:
In this paper, a novel approach is developed to update the background. This approach is based
on a double background principle, long-term background and short-term background. For the
long-term background, the background interference information which has happened in a long
time is saved. For the short-term background, the most recent changes are saved. These two
background images are modified to adequately update the background image and to detect
and correct abnormal conditions. During practical tests, we found that although the optical flow
of background interference can be detected without moving object, it is relatively stable for
some specific areas on the image and the amount of this optical flow doesn't change very
much. For the area where the moving object appears, the amount of optical flow must change

19. significantly in the specific area. According to these characteristics, the moving object should be
easily detected if the information of the background and foreground can be separated. In this
paper, a method entitled Double Background Filtering (DBF) is proposed, which consists of four
steps. Figure.4.6 shown below explains the method in a tabular way.
1. The optical flow information of the first five frames is accumulated for saving the optical
flow information of the background interference. Let 5 A be the accumulation matrix,
which is defined with the same size as the video images and set the initial value as zeros.
To compute this matrix the formula below is applied:
2. The optical flow information of the last three frames is accumulated for moving object
detection. Let 3 A be the accumulation matrix and computed as follow:
3. By comparing the results of steps (1) and (2) and eliminating the overlap optical flow,
the rest should be the optical flow which represents as the real movement. The
algorithm to detect whether a pixel B(i, j) belongs to an object with salient motion is
described as follows
4. Background Updating, This step is an updating function of the new value of the
accumulation matrix, both 5 A and 3 A are set to zero, with the new video frame input,
the four steps above are then repeated.
In this method, there are always two unused frames during the process, the purpose of this is
to separate the background and moving object effectively. When the moving object appears in
the last three frames, the information of moving object will not be lost while the background is
updating. Figure 4.7 shows the result of double background filtering method. From the results,
we can see that for the background without moving object, the background interference can
not be eliminated completely and for the background with moving object, although the moving
object area can be detected, the background interference is still exit. So how to get rid of the
background interference and preserve the information of moving object at the same time is
most important problem we are facing. The morphological processing method is introduced in
next section to solve this problem.
Morphological Image Processing:
Morphological image processing is a collection of techniques for digital image processing based
on mathematical morphology which is a nonlinear approach that is developed based on set
theory and geometry [29]. Morphological image processing techniques are widely used in the
area of image processing, machine vision and pattern recognition due to its robustness in
preserving the main shape while suppressing noise. When acting upon complex shapes,
morphological operations are able to decompose them into meaningful parts and separate
them from the background, as well as preserve the main shape characteristics. Furthermore,

20. the mathematical calculation involved in mathematical morphology includes only addition,
subtraction and maximum and minimum operations without any multiplication and division.
There are two fundamental morphological operations which are dilation and erosion and many
of the morphological algorithms are based on these two primitive operations. Dilation of the
set A by set B which is usually called as structure element, denoted by AÅ B, is obtained
by first reflecting B about its origin and then translating the result by x. All x such that A and
reflected B translated by x that have at least one point in common form the dilated set.
In our experiments, we use three morphological operators, dilation, opening and closing. The
first one, dilation, is applied in the image with the accumulation optical flow for the first five
frames, which is after the first step of DBF method. The dilation operator expands the area of
background interference to make it eliminated efficiently in the third step of DBF method. The
other two operators, opening and closing, are applied in the image with the accumulation
optical flow for last three frames, which is after second step of DBF method. The opening
operator is used first to eliminate the noise which consists of isolated points and closing
operator is used immediately after filling up the holes and gaps. The structure element in both
operations is SE = {1,1,1;1,1,1;1,1,1}. Figure 4.8 shows the results of DBF with morphological
processing. From the results, we can see that the DBF method with morphological processing
can preserve the moving object area very well and eliminate the background interference
completely. The result of this processing can be very helpful for the further motion area
detection.
MOTION AREA DETECTION:
After applying the step of DBF method with morphological processing, the optical flow
information of the background interference should be eliminated and only the optical flow
information of real moving object is left. During the experimental test, we find that the
appearance of a moving object can make a big influence on the instantaneous rate of change
between the foreground motion information and the accumulative background optical flow
information. In this paper, we use the result of DBF method with morphological processing as
the foreground motion information FM. Because the result of DBF method with morphological
processing comes from the last three frames accumulative optical flow information so that the
result of the first seven frames accumulative optical flow information is used as the
accumulative background optical flow information ABOF7. So we can define the instantaneous
rate of change for the moving object appearance IRCA as follows:
From the results, we can see that, for the background without moving object, IRCA has a small
value with little changing. But if there is moving object appearance, the value of IRCA will
increase sharply and last for several frames time. By taking advantage of this feature, we can

21. use this IRCA value to detect the movement of moving object and give the alarm without delay.
In our experiment, the alarm threshold T is set as 0.25 and the abnormity alarm will occur
whenever the IRCA value is above T. It can be described as follows:
CONCLUSION APPLICATIONS AND FUTURE SCOPE
Conclusion:
A video monitoring & detection system was thus developed successfully in this paper. This
system mainly provides an efficient method for surveillance purposes and is aimed to be highly
beneficial for any person or organization. Thus, a motion based change detection in avi video
format was completed and successfully implemented.
Applications:
 Expert Systems
 Home Security Systems
 Industrial Security Systems
 Airports
 Government Buildings
 Research Facilities
 Military Facilities
 Medical Imaging
 Artificial Intelligence
 Micro Robots
 Machine Vision Applications
Future Scope:
The future scope of the work done could be as follows: the due course of time as we started to
understand the minute details of our work, we significantly realized that our software would be
tremendously important in the future world.
Following changes or additions can be done to include some new features:
 With the existing alarm system, advancement can be included and SMS can be sent to
the user when motion is detected.
 The stored video can be automatically transferred to some email account so that an
extra backup data can be used.

22.  A user_id and password can be given to a user so that unauthorized people don't have
access to the software.
 A facility for the user can be given where he can mainly monitor only a small specific
area in the range of the Web cam.
 In the future, the user can be provided a remote access to this software from some
remote PC through internet.
 Include an option to take snaps periodically, manually or automatically.
 Work could be done to make the system more users friendly for a layman user.
REFERENCES:
1. Duane C. Hanselman and Bruce L. Littlefield, "Mastering Matlab 7".
2. www.mathworks.com.
3. www.matlab.com.
4. Rozinet, O. and Z. Szabo, "Hand motion detection using Matlab software environment".
5. Nehme, M.A.; Khoury, W.; Yameen, B.; Al-Alaoui, M.A., "Real time color based motion
detection and tracking", Proc. ISSPIT 2003, 3rd IEEE International Symposium on Signal
Processing and Information Technology, 2003, 14-17 Dec. 2003, pp. 696 - 700, 14-17
Dec. 2003.
6. Josué A. Hernández-García, Héctor Pérez-Meana
and Mariko Nakano-Miyatake, "Video Motion Detection Using the Algorithm of
Discrimination and the Hamming Distance", Lecture Notes in Computer Science,
Springer-Verlag, Germany.
7. H.A.M. El_Salamony, H.F. Ali, and A.A. Darweesh, "3D Human Body Motion Detection
and Tracking in Video", Proc. Acta Press.
8. Song, Y.,"A perceptual approach to human motion detection and labeling", PhD thesis,
California Institute of Technology, 2003.
9. Yilmaz, A., M. Shah, "Contour Based Object Tracking with Occlusion Handling in Video
Acquired Using Mobile Cameras", Proc. IEEE Transactions on Pattern Analysis and
Machine Intelligence, 2005.
10. Borst, A. and Egelhaaf, M., "Principles of visual motion detection", Trends in
Neurocience, Vol. 12, pp. 297-305, 1989.
11. Y.L. Tian and A. Hampapur, "Robust Salient Motion Detection with Complex Background
for Real-time Video Surveillance," IEEE Computer Society Workshop on Motion and
Video Computing, Breckenridge, Colorado, January 5 and 6, 2005.
12. C. Bahlmann, Y. Zhu, Y. Ramesh, M. Pellkofer, T. Koehler, "A system for traffic sign
detection, tracking, and recognition using color, shape, and motion information," IEEE
Intelligent Vehicles Symposium, Proceedings,2005, pp. 255-260.

23. 13. A. Manzanera and J.C. Richefeu, "A new motion detection algorithm based on
background estimation," Pattern Recognition Letters, vol. 28, n 3, Feb 1, 2007, pp. 320-
328.
14. Y, Ren, C.S. Chua and Y.K. Ho, "Motion detection with nonstationary background,"
Machine Vision and Applications, vol. 13, n 5-6, March, 2003, pp. 332-343.
15. J. Guo, D. Rajan and E.S. Chng, "Motion detection with adaptive background and
dynamic thresholds," 2005 Fifth International Conference on Information,
Communications and Signal Processing, 06-09 Dec. 2005, pp. 41-45.
16. A. Elnagar and A. Basu, "Motion detection using background constraints," Pattern
Recognition, vol. 28, n 10, Oct, 1995, pp. 1537-1554.
17. J.F. Vazquez, M. Mazo, J.L. Lazaro, C.A. Luna, J. Urefla, J.J. Garcia and E. Guillan,
"Adaptive threshold for motion detection in outdoor environment using computer
vision," Proceedings of the IEEE International Symposium on Industrial Electronics, 2005.
ISIE 2005,vol. 3, 20-23 June 2005, pp. 1233-1237.
18. P. Spagnolo, T. D'Orazio, M. Leo and A. Distante, "Advances in background updating and
shadow removing for motion detection algorithms," Lecture Notes in Computer Science,
vol. 3691 LNCS, 2005, pp. 398-406.
19. D.E. Butler, V.M. Bove Jr. and S. Sridharan, "Real-time adaptive foreground/background
segmentation," Eurasip Journal on Applied Signal Processing, vol. 2005, n 14, Aug 11,
2005, pp. 2292-2304.
20. Y.S. Choi, Z.J. Piao, S.W. Kim, T.H. Kim and C.B. Park, "Salient motion information
detection technique using weighted subtraction image and motion vector," Proceedings
of 2006 International Conference on Hybrid Information Technology, vol. 1, 2006, pp.
263-269.
21. J.W. Wu and M. Trivedi, "Performance characterization for Gaussian mixture model
based motion detection algorithms," Proceedings of International Conference on Image
Processing, vol. 1, 2005, pp. 1097-1100.
22. P.R.R. Hasanzadeh, A. Shahmirzaie and A.H. Rezaie, "Motion detection using differential
histogram equalization," Proceedings of the Fifth IEEE International Symposium on
Signal Processing and Information Technology, 2005, pp. 186-190.
23. D.X. Zhou and H. Zhang, "Modified GMM background modeling and optical flow for
detection of moving objects," Conference Proceedings of IEEE International Conference
on Systems, Man and Cybernetics, vol. 3, 2005, pp. 2224-2229.
24. G. Jing, C.E. Siong and D. Rajan, "Foreground motion detection by difference-based
spatial temporal entropy image," IEEE Region 10 Conference Proceedings: Analog and
Digital Techniques in Electrical Engineering, 2004, pp. 379-382.
25. C.C. Chang, T.L. Chia and C.K. Yang, "Modified temporal difference method for change
detection," Optical Engineering, vol. 44, n 2, February, 2005.

24. 26. J. Lopez, M. Markel, N. Siddiqi, G. Gebert and J. Evers, "Performance of passive ranging
from image flow," IEEE International Conference on Image Processing, vol. 1, 2003, pp.
929-932.
27. N. Lu, J.H. Wang, Q.H. Wu and L. Yang, "Motion detection based on accumulative optical
flow and double background filtering," Proceedings of World Congress on Engineering,
London, UK, 2-4 July, 2007, pp. 602-607.
28. J. Lin, J.H. Xu, W. Cong, L.L. Zhou and H. Yu, "Research on real-time detection of moving
target using gradient optical flow," IEEE International Conference on Mechatronics and
Automation, 2005, pp. 1796-1801.
29. E. Trucco, T. Tommasini and V, Roberto, "Near-recursive optical flow from weighted
image differences," IEEE Transactions on Systems, Man, and Cybernetics, Part B:
Cybernetics, vol. 35, n 1, February, 2005, pp.124-129.
30. H. Ishiyama, T. Okatani and K. Dequchi, "High-speed and high-precision optical flow
detection for real-time motion segmentation, " Proceedings of the SICE Annual
Conference, 2004, pp. 751-754.
31. L. Wixson, "Detecting salient motion by accumulating directionally-consistent flow," IEEE
Transactions on Pattern Analysis and Machine Intelligence, vol. 22, issue 8, Aug. 2000,
pp. 774-780.
32. J. Barron, D. Fleet and S. Beauchemin, "Performance of Optical Flow Techniques,"
International Journal of Computer Vision, vol. 12, n 1, 1994, pp. 42-77.
33. H. Liu, T. Hong, M. Herman and R. Chellappa, "Accuracy vs. Efficiency Trade-offs in
Optical Flow Algorithms," In the Proceeding of Europe Conference Of Computer Vision,
1996.
34. B. Galvin, B. McCane, K. Novins, D. Mason, and S.Mills, "Recovering Motion Fields: An
Evaluation of Eight Optical Flow Algorithms," In Proc. Of the 9th British Machine Vision
Conference (BMVC'98), vol. 1, Sep. 1998, pp. 195-204.
35. Wikipedia, the free encyclopedia. 20 February 2007. Lucas Kanade method. Available:
http://en.wikipedia.org/wiki/Lucas_Kanade_method
36. Y. Shan and R.S. Wang, "Improved algorithms for motion detection and tracking,"
Optical Engineering, vol. 45, n 6, June 2006.
37. H.J. Elias, O.U. Carlos and S. Jesus, "Detected motion classification with a double-
background and a neighborhood-based difference," Pattern Recognition Letters, vol. 24,
n 12, August, 2003, pp.2079-2092.
38. J. Zheng, B. Li, B. Zhou and W. Li, "Fast motion detection based on accumulative optical
flow and double background model," Lecture Notes in Computer Science,
Computational Intelligence and Security - International Conference, CIS 2005,
Proceedings, vol. 3802 LNAI,2005, pp. 291-296.

25. 39. R.C. Gonzalez and R.E. Woods, Digital Image Processing, Prentice-Hall, December, 2002,
Second edition, ISBN: 0130946508.