Developing A Laptop Intrusion Security System Computer Science Essay

Published:

This project discusses the new security system that will be cheaper, very easy to install and efficient. On the other hand it can be made available to everyone. This system will use a system that is laptop or computer, a web camera that connected to that system. This web camera will take video as input and detect any moving object inside the video. If any moving object is found then system will detect as intrusion. Then a SMS will be sent to the user via the internet and SMS gateway. In this project our main objective is to detect motion in the video captured by camera. For that we will use Java as programming language. Again for capture the video using camera we need another technology. For that we will use Java Media Framework (JMF) to capture video using the camera. Then will use the internet and SMS gateway for sending the alert message to the user.

Lady using a tablet
Lady using a tablet

Professional

Essay Writers

Lady Using Tablet

Get your grade
or your money back

using our Essay Writing Service!

Essay Writing Service

Security system is very important for home or organizations. A good security system can help a home or office secure from intruders. But the problem is the cost. If we want a good security system we will have to expend a lot. That's why most of the time we cannot install the security system though it is required. So it is very important to introduce new security system for home and office that is cheaper and effective as well. We should not sacrifice the performance that we need from a good security system to reduce the cost of the system.

Researchers are trying to invent newer security system using several modern technologies. They are applying several technologies like image processing, face recognition, alarming, etc. But it is not being possible to reduce the production cost. As a result, the security system is still out of reach of the general people. In my project, I am proposing a new security system that will be cheaper, easy to install and efficient as well.

In my system, I will use a desktop or laptop computer, a web camera connected to the computer. The web camera will be capturing the region where we want the security to be implemented. Software installed in the computer will be running all the time. This software will take the video as input and detect any moving object inside the video. If any moving object is found system will detect as intrusion. Then a SMS will be sent to the user via the internet and SMS gateway. The frames will also be saved those are responsible for the motion to view later. The images will be saved in a pre-defined directory.

The system is simple and easy to implement. The rest of the book describes the background study for the motivation of this project, the security system description, algorithm, and system architecture and system flowchart. This chapter also provides the pros and cons of the system with reason behind them. Finally a chapter concludes the book. The book also provides appendices and bibliography used to develop the system.

The organization of this book is as follows:

The chapter 2 provides the background study of my project work. Our main objective is to detect motion in video captured by camera. This chapter gives a clear idea on the motion detection. To capture the video using camera we need another technology. There are lots of technologies for that. We used Java Media Framework (JMF) to capture video using the camera. The starting guideline for JMF is also given in this chapter.

The chapter 3 is all about our developed system. This chapter describes the requirements of the system. System description, algorithm, architecture and flowchart are given in this chapter. This chapter also explains the benefits and limitations of this system.

The chapter 4 is the concluding chapter of this book.

Appendices and references are given at the end of the book.

Chapter 2

Literature Review

This chapter discusses the literature reviewed to develop the Simple SMS Based Security

System. Our main goal is to detect motion from a video. For defining activity in a scene by analyzing image data and differences in a series of images, and a way to this job is Video Motion Detection (VMD). We used software that uses a motion detection algorithm to detect motion and send SMS to a particular user of this system. We used Java as programming platform and Java Media Framework (JMF) to capture video and working on it.

Lady using a tablet
Lady using a tablet

Comprehensive

Writing Services

Lady Using Tablet

Plagiarism-free
Always on Time

Marked to Standard

Order Now

This chapter provides brief idea on several motion detection algorithms and their pros and cons in section 2.1 and its sub-sections. Finally, this chapter explains the algorithms that we have selected for our project in the section 2.2. This chapter also provides some information on Java Media Framework (JMF) in part 2.3. And part 2.4 is the conclusion of the chapter.

2.1 Motion Detection

Almost every visual surveillance system establishes with motion detection. Motion detection means segmenting regions corresponding to moving objects from the remaining of an image. Consequent processes for instance tracking and behavior recognition are very much dependent on it. The procedure of motion detection typically engages environment modeling, motion segmentation, and object classification, which intersect each other during processing.

2.1.1 Environment Modeling

The active construction and updating of environmental models are very important to visual surveillance. Environmental models can be classified into 2-D models in the image plane and 3-D models in real world coordinates. Due to their simplicity, 2-D models have more applications.

For fixed cameras, the key problem is to involuntarily make progress and modify background images from a dynamic sequence. Unfavorable factors, such as illumination variance, shadows and shaking branches, bring many difficulties to the achievement and updating of background images. There are several algorithms for resolving these problems together with temporal average of an image sequence [8], [12], adaptive Gaussian estimation [8], and parameter estimation based on pixel processes [16], [17], etc. Ridder et al. [18] model each pixel value with a Kalman Filter to compensate for illumination variance. Stauffer et al. [5], [17] present a theoretic framework for recovering and updating background images based on a process in which a mixed Gaussian model is used for each pixel value and online estimation is used to update background images in order to adapt to illumination variance and disturbance in backgrounds. Toyama et al. [20] propose theWallflower algorithm in which background maintenance and background subtraction are carried out at three levels: the pixel level, the region level, and the frame level. Haritaoglu et al. [2] build a statistical model by representing each pixel with three values: its minimum and maximum intensity values and the maximum intensity difference between successive frames experimented throughout the training period. These three values are updated periodically. An adaptive background model is used in McKenna et al. [4] by color and gradient information to reduce the influences of shadows and unreliable color cues.

For pure translation (PT) cameras, an environment model can be made by patching up a panorama graph to acquire a holistic background image [21]. Homography matrices can be used to describe the transformation relationship between different images.

For mobile cameras, motion compensation is needed to construct temporary background images [22]. Regarding 3-D environmental models [23], current work is still limited to indoor scenes because of the difficulty of 3-D reconstructions of outdoor scenes.

2.1.2 Motion Segmentation

Motion segmentation in image sequences aims at detecting regions corresponding to moving objects such as vehicles and humans. Detecting moving areas provides a focus of concentration for later procedures such as tracking and behavior analysis because only these regions required to be considered in the later procedures. At present, most segmentation methods use either temporal or spatial information in the image sequence. Several conventional approaches for motion segmentation are outlined in the following.

2.1.2.1. Background subtraction.

Background subtraction is a popular method for motion segmentation, especially under those situations with a relatively static background. It detects moving regions in an image by taking the difference between the current image and the reference background image in a pixel-by-pixel fashion. It is simple, but extremely sensitive to changes in dynamic scenes derived from lighting and extraneous events etc. Therefore, it is highly dependent on a good background model to reduce the influence of these changes [2], [4], [5], as part of environment modeling.

2.1.2.2 Temporal differencing.

Temporal differencing makes use of the pixel-wise differences between two or three consecutive frames in an image sequence to extract moving regions. Temporal differencing is very adaptive to dynamic environments, but generally does a poor job of extracting all the relevant pixels, e.g., there may be holes left inside moving entities. As an example of this method, Lipton et al. [3] detect moving targets in real video streams using temporal differencing. After the absolute difference between the current and the previous frame is obtained, a threshold function is used to determine changes. By using a connected component analysis, the extracted moving sections are clustered into motion regions. An improved version uses three-frame instead of two-frame differencing.

2.1.2.3. Optical flow.

Lady using a tablet
Lady using a tablet

This Essay is

a Student's Work

Lady Using Tablet

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Examples of our work

Optical-flow-based motion segmentation uses characteristics of flow vectors of moving objects over time to detect moving regions in an image sequence. For example, Meyer et al. [6], [14] compute the displacement vector field to initialize a contour based tracking algorithm, called active rays, for the extraction of articulated objects. The results are used for gait analysis. Optical-flow-based methods can be used to sense separately moving objects even in the occurrence of camera motion. However, most flow computation methods are computationally complex in addition to very susceptible to noise, and cannot be applied to video streams in real time without specialized hardware. More detailed discussion of optical flow are described in Barron's work [7].

Of course, besides the basic methods described above, there are some other approaches for motion segmentation. Using the extended expectation maximization (EM) algorithm, Friedman et al. [8] implement a mixed Gaussian classification model for each pixel. This model classifies the pixel values into three separate predetermined distributions corresponding to background, foreground and shadow. It also updates the mixed component automatically for each class according to the likelihood of membership. Hence, slowly moving objects are handled perfectly, while shadows are eliminated much more effectively. VSAM [1] has successfully developed a hybrid algorithm for motion segmentation by combining an adaptive background subtraction algorithm with a three-frame differencing technique. This hybrid algorithm is very fast and surprisingly effective for detecting moving objects in image sequences. In addition, Stringa [9] proposes a novel morphological algorithm for detecting motion in scenes. This algorithm obtains stable segmentation results even under varying environmental conditions.

2.1.3. Object Classification

Different moving regions may correspond to different moving targets in natural scenes. For instance, the image sequences captured by surveillance cameras mounted in road traffic scenes probably include humans, vehicles and other moving objects such as flying birds and moving clouds, etc. To further identify objects and investigate their behaviors, it is essential to correctly classify moving objects. Object classification can be thought as a standard pattern recognition topic. At present, there are two main categories of approaches for classifying moving objects.

2.1.3.1 Shape-based classification.

Different descriptions of shape information of motion regions such as points, boxes, silhouettes and blobs exist to group moving objects. In the paper [1], image blob dispersedness, image blob area, apparent aspect ratio of the blob bounding box as main features are taken. This paper also classifies moving-object blobs into four classes: single human, vehicles, group of human and clutter with a viewpoint precise three-layer neural network classifier. The area of image blobs and the dispersedness is used in Lipton et al. [3] as classification metrics to classify all moving-object blobs into humans, vehicles and clutter. Chronological consistency constraints are taken under consideration to make classification results more precise. The simple shape parameters of human silhouette models are used in Kuno et al. [10] to separate humans from other moving objects.

Motion-based classification.

In general, non-rigid articulated human motion give an idea about a cyclic property, so this has been used as a strong indication to categorize moving objects. A similarity-based technique to detect and analyze periodic motion is described in Cutler et al. [11]. By tracking a remarkable moving object its self correspondence is computed as it evolves over time. As we know a for cyclic motion, its self-correspondence measure is also cyclic. Hence time-frequency study is applied to identify and characterize the periodic motion, and tracking and classification of moving objects are executed using periodicity. Residual flow is used in Lipton's work [12] to analyze firmness and periodicity of moving objects. It is expected that rigid objects present little residual flow, where as a non-rigid moving object for instance a human being has a more average residual flow and even display a periodic component. Human motion is distinguished based on this useful cue from motion of other objects, such as vehicles.

The two general approaches stated above, specifically shape-based and motion-based classification can also be efficiently combined for classification of moving objects. Additionally, Stauffer [13] proposes a new technique standing on a time co-occurrence matrix to hierarchically classify both objects and behaviors. Obtaining more precise classification results by using extra features such as color and velocity is expected.

2.2 Temporal differencing to Detect Motion Regions

From the section stated above, we have found three popular motion area detection methods. These are background subtraction, temporal differencing and optical flow. Background subtraction is a popular method for motion segmentation, especially under those situations with a relatively static background. It detects moving regions in an image by taking the difference between the current image and the reference background image in a pixel-by-pixel fashion. It is simple, but extremely sensitive to changes in dynamic scenes derived from lighting and off the point events etc. Therefore, it is highly dependent on a good background model to reduce the influence of these changes [2], [4], [5], as part of environment modeling. Again, optical-flow-based motion segmentation uses characteristics of flow vectors of moving objects over time to detect moving regions in an image sequence. For example, Meyer et al. [6], [14] compute the displacement vector field to initialize a contour based tracking algorithm, called active rays, for the extraction of articulated objects. The results are used for gait analysis. Optical-flow-based methods can be used to sense separately moving objects even in the occurrence of camera motion. However, most flow computation methods are computationally complex in addition to very susceptible to noise, and cannot be applied to video streams in real time without specialized hardware. Temporal differencing makes use of the pixel-wise differences between two or three consecutive frames in an image sequence to extract moving regions. Temporal differencing is very adaptive to dynamic environments. As an example of this method, Lipton et al. [3] detect moving targets in real video streams using temporal differencing. After the absolute difference between the current and the previous frame is obtained, a threshold function is used to determine changes. By using a connected component analysis, the extracted moving sections are clustered into motion regions. An improved version uses three-frame instead of two-frame differencing.

Temporal differencing algorithm is simple to put into action. Therefore, temporal differencing algorithm is selected for detecting motion in our project. The algorithm is described in the following subsection.

2.2.1 The Algorithm

There are many variations in temporal differencing method. The simplest is to take consecutive video frames and determine the absolute difference. In the paper [3], a threshold function is used to determine the change. If is the intensity of the nth frame, then the pixel wise difference function is:

Then a motion image can be extracted by threshold values as follows:

Where, u and v is the row and column number of the pixel and T is the threshold value. If is greater or equal to the threshold value (T), motion is detected at pixel (u,v) of nth frame. If is lower than the threshold value (T), motion is not detected at pixel (u,v) of nth frame.

The threshold T has been determined empirically to be ≈ 15% of the digitizer's brightness range. For a digitizer providing 255 grey levels, a value of T ≈ 40 should be used.

2.3. Java Media Framework

The Java Media Framework (JMF) is an application programming interface (API) for putting together time based media into applications developed by Java and Java applets [A]. JMF 2.0 is premeditated for live support of capturing media data and easy program. It enables the development of media streaming and conferencing applications in Java. It also enables advanced developers and technology contributors to put into practice the custom solutions based on the on hand API and easily integrate new features with the existing framework.

The Java Media Framework (JMF) offers a number of previously built classes that handle the reading, processing and display of data.  A media can simply be included into any graphical application developed either with AWT or Swing, with the Player class.  The Processor permits you to manage the encoding or decoding process at a improved level than the Player, such as adding a custom codec or effect among the inputs and outputs.

To embed multimedia in an application the Player class can play a vital role.  It can be an easy solution to do this. The setup of the file handler, video and audio decoders, and media renderers automatically are handled by this Player class.  It is possible to embed the Player in a Java application developed using Swing framework. But to use the Player class in any application, care must be taken as it is a heavy weight component.

2.4. Conclusion

We have studied several motion area segmentation algorithms. Among them we have found three popular algorithms. They are background subtraction, temporal differencing and optical flow. Background subtraction is effective if the background is static. It is simple, but extremely sensitive to changes in dynamic scenes derived from lighting and off the point events etc. Therefore, this method can not give good result in most of the cases. Optical-flow-based methods can be used to sense separately moving objects even in the occurrence of camera motion. However, most flow computation methods are computationally complex in addition to very susceptible to noise and cannot be applied to video streams concurrently with no dedicated hardware. On the other hand, Temporal differencing makes use of the pixel-wise dissimilarities among two or three successive frames in an image progression to extract moving regions. Temporal differencing is very adaptive to dynamic environments.

The Java Programming Language is a very powerful language to develop systems like our image processing system. Java Media Framework is a very good media processing tool now a day. Combining these two will make the system easy to develop and implement.

Chapter 3

A Simple SMS Based Security System

This chapter discusses the details of the simple SMS based security system. The section 3.1 describes the requirement analysis of the system. The section 3.2 gives the overview of the security system. This section provides the system description in subsection 3.2.1. The 3.2.2 explains the algorithm. The architecture of the system is described in 3.2.3. In this section, the function of each part in the system architecture will be described. The system flowchart is given in the section 3.2.4. The benefits achieved from this system are explained in 3.3. The limitations of the system will be given in section 3.4 and section 3.5 concludes the chapter.

3.1 Requirement Analysis of the Simple SMS Based Security System

The system is developed using Java programming language. The Java Development kit is required to compile the programs and Java Virtual Machine (JVM) is required to run the programs. The Java Media Framework is required to control the webcam to capture the video and send to the system for further processing. Java Mail API is required to send SMS through SMS gateway. The system requires a webcam to take video input to the system. We need internet and SMS credit from SMS gateway service provider to send SMS through internet. The following technologies are required for the system:

Software:

Java Development Kit (JDK) version 4.0 or higher

Java Media Framework (JMF) version 2.0 or higher

Java Mail API

Hardware:

Webcam or CC Camera

Personal Computer

Others

Internet

SMS credit from SMS Gateway service provider to send SMS

Overview of the Simple SMS Based Security System

The simple SMS based security system is developed on the basis of motion detection using the concept of comparing frames of a video. The system is developed using Java programming language. When a user of this system runs it, the system will on a camera connected to the computer. The camera captures video and continuously sends the video to the computer. The system takes the video and compares the consecutive frames. If any change is found in pixels of the consecutive video frames, the motion is detected. The system draws rectangles on the pixels those are changed. The motion can be classified as very strong motion, strong motion, weak motion and very weak motion. The red rectangular boxes are drawn if very strong motion is found. If the motion is strong the pink rectangular boxes are drawn. Yellow rectangular boxes are drawn if weak motion is detected. Finally, white rectangular boxes are drawn if very weak motion is detected. As soon as the motion is detected, a SMS is sent to user's cell phone. The system also takes the snaps as soon as the motion is detected and stores them to the computer hard disk. The saved images can be viewed later.

The rest of the texts of this section will narrate the system description, the algorithm on which the system is developed, the system architecture and the system flowchart.

System Description

At first connect the webcam to the computer and check the internet connection. To run the systems first go the command prompt. Using the command prompt go to the project directory. Then run the Main Class. The program will run and show the main window of the SMS Based Security System on the screen. Figure-3(a) shows the main.

Figure-3(a): The Main Window of SMS Based Security System

There are two buttons in the window. One is START button and another is EXIT button. By clicking the EXIT button system will be closed. By clicking the START button a new window will be opened and the video capturing will be started. The figure-3(b) shows the window.

Figure-3(b): Capturing Video

Initially there is no moving object in the frame of the camera. If anybody comes into the frame of the camera, a motion will be detected and the pixels those are affected due to motion are highlighted by colored rectangular boxes (Figure-3(b)).

Figure-3(c): The motion is detected

Figure-3(d): Alarm Control Panel Reset System Window after the motion is detected.

As soon as the motion is detected, a SMS will be sent to the user and an alarm control panel window will be visible to resume the system. Figure-3(d) shows the window. When user press the resume button the system will resume.

The system is also capable of saving frames as images in a predefined directory. As soon as the motion is detected, the images are saved in the directory. User can see the images later.

The Algorithm

Start the system and prepare the system to take the video in RGB (Red, Green and Blue) format.

Set RGB as the output format.

Set the areas in pixel those need to be compared. It can be said as INITIAL_SQUARE_SIZE. Initially, it is 40.

Set several threshold values for several motion types. Set, 50 for very strong motion, 40 for strong motion, 30 for weak motion and 20 for very weak motion.

Set the colors to highlight the pixels those are changed due to motion. Red for very strong motion, Pink for strong motion, Yellow for the weak motion and white for the very weak motion.

Connect to the camera.

Capture the RGB video and store the video data into the input buffer. Use Java Media Framework (JMF) [24] to do this.

Copy all the data from the input buffer to the output buffer. The purpose is to display the video input on the screen.

Now, simplify the image to black and white. As a result of this, image information shrinks to the one third of the original amount. Hence, less processing is needed.

Now, apply the Temporal Differencing Method specified in the paper [3].

If the motion is detected, the affected pixels are colored based on the threshold value stated above. The current frame and its consecutive frames are saved to a predefined directory. An ALERT message is sent to the user through SMS. SMS is sent using Java MAIL API [25].

An "Alarm Control Panel" is shown. If the system is resumed, then continue. Otherwise, the system is paused.

If motion is not detected, then repeat the steps from 7 to 12.

Exit.

3.2.3 System Architecture

The system has four classes. The classes are- Main, SimpleApp, MotionProcessor and ControlPanel. The figure-3(e) shows the class diagram for the system. The functions of the classes are given bellow:

Main- This class is builds the main GUI. This is the class that shows the main panel of the System interface. There are two buttons in the window. One is START button and another is EXIT button. By clicking the EXIT button system will be closed. By clicking the START button, an object of the SimpleApp class is created and a new window will be opened. Then the video capturing will be started. The figure-3(b) shows the window.

SimpleApp- Creates a processor and put the processor into the configured state so that it can be used as a player. After that it obtains the track controls and searches for the track control for the video track. Then it instantiates and sets the frame access codec to the data flow path. After that the processor is started and the system is blocked until the processor has transitioned to the given state. It returns false if the transition failed.

MotionProcessor - This class sets the input and output format. This class first scales down the image. Then converts the image to an int[][] array instead of using byte[][]. Then it does all the calculation on int[][] array instead of masking the bit. Then it compares the last frame with the new frame using temporal differencing according to the algorithm specified in the paper [3]. It can detect the pixels those were changed because of motion. It changes the color of only the pixels which changed; the rest is discarded on the black and white image which is used for determining the motion. After the motion is detected, an object the ControlPanel class is created. This object sends a SMS to the user. MotionProcessor also saves consecutive frames to a predefined directory to view later.

Main

SimpleApp

MotionProcessor

ControlPanel

Figure-3(e): Class Diagram of SMS Based Security System

ControlPanel- This class provides a window that contains a reset button to resume the system action after the motion is detected.

3.2.4 System Flowchart

Figure-3(f): Flow Chart of SMS Based Security System

3.3 Benefits of the Simple SMS Based Security System

The Simple SMS Based Security System is very simple as its name implies. The system is developed using Temporal Differencing [3]. Temporal differencing makes use of the pixel-wise differences between two or three consecutive frames in an image sequence to extract moving regions. Temporal differencing is very adaptive to dynamic environments.

The system first converts the original picture and reference picture to black and white and then calculates the difference in pixel. As the image is converted to black and white image before comparing, it requires less computation. Hence, the system requires less computing resources.

The system required a personal computer and a web camera. The system uses SMS gateway to send SMS to the user. The system is connected to the SMS gateway through internet. So, we need no cell phone to be connected with the computer. Hence, this system is not dependent on the cell phone model to send SMS. The processing ability required for the personal computer is not so high. We can use usual personal computer to develop the system. So, it is very easy to effort such kind of security system. The system also saves images after the motion is detected. The images are stored in computer hard disk drive. These images can be viewed later to have clear ideas about the alerting SMS.

3.4 Limitations

Though the system is simple to implement, it has some computational limitations. The system is developed on the basis of motion detection using the temporal differencing algorithm [3]. Temporal differencing makes use of the pixel-wise differences between two or three consecutive frames in an image sequence to extract moving regions. Temporal differencing is very adaptive to dynamic environments, but generally does a poor job of extracting all the relevant pixels, e.g., there may be holes left inside moving entities. The system first converts the original picture and reference picture to black and white and then calculates the difference in pixel. This results in loss of image information. Adequate lighting is required to have a good snap. The system does not work efficiently in low light. Again another important issue is the camera resolution. If the resolution is not so good, the quality of the image captured by the camera will not be good. If the image quality is not good enough the motion may not be detected properly. The system motion detection depends on a threshold value as stated earlier. If the threshold value is lower, the change in pixels will be higher. As a result the motion may be detected frequently. If the threshold value is higher, the motion may be detected by no means. It is very difficult to set a value for the threshold. The system cannot be used without internet connection.

3.5 Conclusion

Though it's a simple security system, it has some benefits. Anybody who owns a personal computer and internet connection can use the system. The system requires high resolution camera to have good quality image to compare. It also requires adequate lighting to work efficiently.

Chapter 4

Conclusion

Security system is very important for home or organizations. A good security system can help a home or office secure from intruders. But the problem is the cost. If we want a good security system we will have to expend a lot. That's why most of the time we cannot install the security system though it is required. So it is very important to introduce new security system for home and office that is cheaper and effective as well. We should not sacrifice the performance that we need from a good security system to reduce the cost of the system.

The Simple SMS Based Security System is very simple as its name implies. The system is developed using Temporal Differencing [3]. Temporal differencing makes use of the pixel-wise differences between two or three consecutive frames in an image sequence to extract moving regions. Temporal differencing is very adaptive to dynamic environments.

The system first converts the original picture and reference picture to black and white and then calculates the difference in pixel. As the image is converted to black and white image before comparing, it requires less computation. Hence, the system requires less computing resources.

The system required a personal computer and a web camera. The system uses SMS gateway to send SMS to the user. The system is connected to the SMS gateway through internet. So, we need no cell phone to be connected with the computer. Hence, this system is not dependent on the cell phone model to send SMS. The processing ability required for the personal computer is not so high. We can use usual personal computer to develop the system. So, it is very easy to effort such kind of security system. The system also saves images after the motion is detected. The images are stored in computer hard disk drive. These images can be viewed later to have clear ideas about the alerting SMS.

Though the system is simple to implement, it has some computational limitations. The system is developed on the basis of motion detection using the temporal differencing algorithm [3]. Temporal differencing makes use of the pixel-wise differences between two or three consecutive frames in an image sequence to extract moving regions. Temporal differencing is very adaptive to dynamic environments, but generally does a poor job of extracting all the relevant pixels, e.g., there may be holes left inside moving entities. The system first converts the original picture and reference picture to black and white and then calculates the difference in pixel. This results in loss of image information. Adequate lighting is required to have a good snap. The system does not work efficiently in low light. Again another important issue is the camera resolution. If the resolution is not so good, the quality of the image captured by the camera will not be good. If the image quality is not good enough the motion may not be detected properly. The system motion detection depends on a threshold value as stated earlier. If the threshold value is lower, the change in pixels will be higher. As a result the motion may be detected frequently. If the threshold value is higher, the motion may be detected by no means. It is very difficult to set a value for the threshold. The system cannot be used without internet connection.

The performance of the system can be improved by making a little change to the motion detection technique. If the frames are converted to grayscale or kept in RGB format rather converting it to black and white images, the change in motion will be detected clearly. Again, the system can be improved by including object tracking feature.

Despite the fact that it's a simple security system, it has some benefits. Anybody who owns a personal computer and internet connection can use the system. On the other hand it is very easy to use and install implement as well. Cost effective and possible to make available to everyone.