Scribd
Upload
40 views12 pages

K.S.R College of Engineering TIRUCHENGODE-607215

This document summarizes a student project on video analysis using background subtraction. It presents a new method using Dirichlet process Gaussian mixture models to estimate per-pixel background distributions. This allows the number of modes to be automatically determined, avoiding over-fitting or under-fitting of data. The system is intended for a low-cost intelligent wireless security and monitoring solution that detects moving objects and notifies a central control unit via SMS when a threshold is reached. It was developed using Visual Studio with a VB and C# frontend and SQL Server backend on Windows OS.

Uploaded by

thebhas1954
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
40 views12 pages

K.S.R College of Engineering TIRUCHENGODE-607215

This document summarizes a student project on video analysis using background subtraction. It presents a new method using Dirichlet process Gaussian mixture models to estimate per-pixel background distributions. This allows the number of modes to be automatically determined, avoiding over-fitting or under-fitting of data. The system is intended for a low-cost intelligent wireless security and monitoring solution that detects moving objects and notifies a central control unit via SMS when a threshold is reached. It was developed using Visual Studio with a VB and C# frontend and SQL Server backend on Windows OS.

Uploaded by

thebhas1954
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd

K.S.

R COLLEGE OF ENGINEERING
TIRUCHENGODE-607215.
DEPARTMENT OF
COMPUTER SCIENCE & ENGINEERING

Video Analysis by Background Subtraction


By
[Link]

731511104047

[Link]

731511104045

[Link] kumar

731511104020

[Link]

731511104052

Under the guidance :


[Link] Sujai M.E(Asst. prof.)

ABSTRACT

Video analysis often begins with background subtraction,


we present a new method based on Dirichlet process.
Gaussian mixture models, which are used to estimate
per-pixel background distributions.
Using a non-parametric Bayesian method allows perpixel mode counts to be automatically inferred, avoiding
over-under- fitting.
It results in detecting moving objects provides a focus of
attention for recognition, classification, and activity
analysis.

EXISTING SYSTEM
Smart surveillance cameras with integrated visual
processing modules are presented.
The smart surveillance camera would store all the
videos. But it would take more time to view and capture
the theft due to rewinding of video.
Due to continuous storing of raw data of video, the
threshold increases. So that alarming will get delayed.

DISADVANTAGE
Its difficult to maintain heavy amount of raw video data.
Require higher bandwidth for transmitting the visual
data.
Slow adapting background models cannot quickly
overcome large changes in the image background.
Where many background pixels are incorrectly classified
as foreground pixels.

PROPOSED SYSTEM

Intelligent video surveillance systems deal with the


monitoring of persistent and transient objects within a
specific environment.
A low-cost intelligent wireless security and
monitoring solution using moving object recognition
technology is presented in this project.

PROPOSED SYSTEM (Continued..)


The subtraction of the current captured image and the
background reaches a certain threshold, a moving
object is thought to be in the current view.
And the GSM Modem will automatically notify the
central control unit or the user through SMS/MMS.

TECHNICAL REQUIREMENTS
Software Requirements:
Platform

- Windows XP/7 32 bit OS

Framework

- Visual Studio 4.0

Front end

- VB with C#.net

Back end

- SQL server 2008

Hardware Requirements:
Pentium i2 processor or later.
Hard disk space of 10 GB or more.
RAM capacity of 128mb or more.
Web camera
GSM

SYSTEM ARCHITECTURE

ADVANTAGES
Classical approaches to motion detection are based on
background subtraction.
We estimate the speed of texture change.
Decide region of foreground object.
Track features.

APPLICATION
Automated surveillance
Action recognition
Intelligent environment
Motion analysis

REFERENCE
Referred from the base paper Background
subtraction with Dirichlet Process Mixture Models.

Referred from books such as Evaluation of


Background Subtraction Techniques for Video
Surveillance.

THANK YOU

You might also like