Lightweight Classroom Surveillance System With MIS Integration
Lightweight Classroom Surveillance System With MIS Integration
Lightweight Classroom Surveillance System With MIS Integration
I. INTRODUCTION
Visual surveillance is a general framework that groups a
number of different computer vision tasks aiming to detect
and track the image sequences and on the next level to retain Fig1: Circuit design with IR sensors.
the images which has motion in it. The ultimate goal in
designing smart visual surveillance systems is to replace the
existing passive surveillance and to minimize the disk space. II. RELATED WORK
In the proposed system, we integrate motion detection to the
Video shot boundary detection has been deeply studied in
existing surveillance system. In the proposed system, CCTV
recent years and has found applications in different domains
camera can be replaced by webcam. The images from the
like video indexing, video compression, video access and
camera are taken for detecting motion. This is done by
others. In the frame work of video indexing tool based on the
comparing the current frame with the previous frame, only the
MPEG7 content description standard, a fast and automatic
frames which are different are considered. The motions in the
temporal segmentation of video content is required in order to
frames are identified based on the threshold value. If the
produce an accurate content description of each temporally
difference count between the frames is more than the
homogenous shot.
threshold value, then the movement is considered as the major
Video shot boundary detection algorithm have to challenge
movement & only those frames will be selected.
the difficulty of finding shot boundary in the presence of
The selected frames are converted to the video file and are
camera and object motion and illumination variations.
sent to the centralized server. At the centralized server the
Moreover different video shot boundaries may present
administrator can view the videos. The proposed method is
different appearances like abrupt temporal changes or smooth
well-suited for modern video-surveillance architectures,
temporal transitions.
where limited storage capacity is in need. Proposed system
The algorithm used is that Video shot boundary detection,
also integrated with Object counter equipped at class room
which segments a video by detecting boundaries between
entrance to count attendance based on each class session,
camera shots. It is focusing on detecting both abrupt and
gradual transitions. If the object is moving smoothly we'll
SACHIN B M, B.E Student, Department of CSE, NIE, Mysore, receive small changes from frame to frame. So, it's impossible
Karnataka, India to get the whole moving object. Things become worse, when
SOURABH S P, B.E Student , Department of CSE, NIE, Mysore, the object is moving so slowly.
Karnataka, India
SUHAS G R, B.E Student , Department of CSE, NIE, Mysore,
Karnataka, India
SHEETAL RAJ G, B.E Student, Department of CSE, NIE, Mysore, III. PROBLEM STATEMENT
Karnataka, India
Ms. PRATHIBHA B S, Assistant Professor, Department of ISE, NIE, The problem here is that, usually the cc camera that kept
Mysore, Karnataka, India inside the classroom keeps recording the things even when
458 www.erpublication.org
Lightweight Classroom Surveillance System With MIS Integration
IV. SOLUTION
Solution for the above problem is making the closed circuit
camera to automatically store the frames which show any kind
of motion in the main server, on requirement the frames are
stitched into one continuous video file and the frames with no Fig 2: IR Sensors (emitters and detectors)
motion are discarded. Once the video file is created the frames
of the same file can be deleted to save memory.
VI. WORKING
Motion detection using background subtraction algorithm is We are using two computers where one is used as a client
used in this project. It's possible to compare the current frame (host) and other as server.
not with the previous one but with the first frame in the video A client will keep track of the count of number of students
sequence. So, if there were no objects in the initial frame, entering the class with the help of IR sensors which is placed
comparison of the current frame with the first one will give us at the entrance. The client system is also connected to a
the whole moving object independently of its motion speed. webcam which is used to monitor the activities going on in the
classroom by capturing it frame by frame. At the client we'll
be maintaining the timetable where count keeps on updating
V. TOOLS for a particular class and will automatically update the same
Microsoft Visual Studio value to the next class if number of students remain the same,
Microsoft Visual Studio is an integrated development otherwise the new count value will be updated.
environment (IDE) from Microsoft. It is used to Server is the system which is connected to the client through
develop computer programs for Microsoft Windows, as well NFS mapping and does the task of converting the frames
as web sites, web applications and web services. Visual received, to a video which can be viewed the Head of the
Studio uses Microsoft software development platforms such Department at any point of time by logging into the server.
as Windows API, Windows Forms, Windows Presentation The process of motion detection happens in such a way that
Foundation, Windows Store and Microsoft Silver light. It can each frame is compared with the first frame in the video
produce both native code and managed code. sequence and all those frames where motion is detected are
Visual Studio supports different programming and allows the saved, while other frames are discarded. These frames are
code editor and debugger to support nearly any programming then transferred to the server by using an NFS technology.
language, provided a language-specific service exists.
Built-in languages include C, C++ and C++/ The Network File System (NFS) is a client/server application
CLI (via Visual), VB.NET (via Visual Basic that lets a computer user view and optionally store and update
.NET), C# (via Visual C#), and F#. Webcam A webcam is a file on a remote computer as though they were on the user's
video camera that streams its image in real time to or through own computer. The user's system needs to have an NFS client
a computer to computer network. When "captured" by and the other computer needs the NFS server.
the computer, the video stream may be saved, viewed or sent
on to other networks via systems such as the internet, and Both of them require that you also have TCP/IP installed
email as an attachment. When sent to a remote location, the since the NFS server and client use TCP/IP as the program
video stream may be saved, viewed or on sent there. Unlike that sends the files and updates back and forth. Its protocol
an IP camera (which connects using Ethernet or Wi-Fi), a uses the Remote Procedure Call (RPC) method of
webcam is generally connected by a USB cable, or similar communication between computers.
cable, or built into computer hardware, such as laptops.
Infrared Sensors Once the frames are received at the server they are converted
IR Sensors work by using a specific light sensor to detect a into a video using Windows Media API or AVI. The Head of
select light wavelength in the Infra-Red (IR) spectrum. By the Department will be having the privilege of seeing the
using an LED which produces light at the same wavelength as particular day's video and it is also possible to view the
what the sensor is looking for, you can look at the intensity of History as well.
the received light. When an object is close to the sensor, the
light from the LED bounces off the object and into the light A detailed report of the count of students in each class will be
sensor. This results in a large jump in the intensity, which we made and sent to concerned authority when it required or on a
already know can be detected using a threshold. weekly basis.
459 www.erpublication.org
International Journal of Engineering and Technical Research (IJETR)
ISSN: 2321-0869, Volume-3, Issue-5, May 2015
With RFID sensors or Bio-metric, we can integrate the
attendance system and take note of who all are attending the
lecture and not just the number of people attending it.
REFERENCES
http://ieeexplore.ieee.org/Xplore/home.jsp.
NFS ILLUSTRATED by Brent Callaghan.
C# 4.5 and .NET 4.0 by Andrew Troelsen.
VII. ADVANTAGES
SACHIN B M, B.E Student, Department of CSE, NIE, Mysore,
It minimizes the recorded file size by discarding video Karnataka, India
frames where there is no human activities, thus the SOURABH S P, B.E Student , Department of CSE, NIE, Mysore,
server space will be reduced. Karnataka, India
SUHAS G R, B.E Student , Department of CSE, NIE, Mysore,
Frames sent through the network can be viewed by the Karnataka, India
administrator as videos. SHEETAL RAJ G, B.E Student, Department of CSE, NIE, Mysore,
No manual effort is required.
Karnataka, India
Ms. PRATHIBHA B S, Assistant Professor, Department of ISE, NIE,
Cost effective. Mysore, Karnataka, India
460 www.erpublication.org