Autonomous Real-time Surveillance System with Distributed IP Cameras

Abstract

An autonomous Internet Protocol (IP) camera based object tracking and behaviour identification system, capable of running in real-time on an embedded system with limited memory and processing power is presented in this Project. The main contribution of this work is the integration of processor intensive image processing algorithms on an embedded platform capable of running at real-time for monitoring the behaviour of pedestrians. The Algorithm Based Object Recognition and Tracking (ABORAT) system architecture presented here was developed on an Intel PXA270-based development board clocked at 520 MHz. The platform was connected to a commercial stationary IP-based camera in a remote monitoring station for intelligent image processing. The system is capable of detecting moving objects and their shadows in a complex environment with varying lighting intensity and moving foliage. Objects moving close to each other are also detected to extract their trajectories which are then fed into an unsupervised neural network for autonomous classification. The novel intelligent video system presented is also capable of performing simple analytic functions such as tracking and generating alerts when objects enter/leave regions or cross tripwires superimposed on live video by the operator.

 

GOAL OF PROJECT

The ability to extract moving objects in real time from live video data using an embedded processor is our primary aim. Alert Sound.

ANALYSIS ON EXISTING SYSTEM

Such surveillance systems are often comprised of black and white, poor quality

analogue videos with little or no signal processing, recorded on the same cassette. Most of the recorded images are of insufficient quality to hold as evidence in a law court. It is also expensive to have human operators monitoring real-time camera footage 24/7. The effectiveness and response of the operator is largely dependant on his/her vigilance rather than the technological capabilities of the surveillance system. Events and activities can be missed, should the concentration level of the operator drop; attentional levels drop significantly after 15 minutes of inactivity in the scene.

PROBLEM DEFINITION

The detection, matching and classification of human appearance is a challenging problem. A further weakness of video detection is the limitation of conventional camera systems to operate under wide dynamic range lighting, which is typical for outdoor applications. Therefore, real-time video based tracking application are mostly constrained with limited.

Disadvantage

Most of the recorded images are of insufficient quality to hold as evidence in a law court

It is also expensive to have human operators monitoring real-time camera footage 24/7.

IDEA ON PROPOSED SYSTEM

The Algorithm Based Object Recognition and Tracking (ABORAT) system presented in this paper is a vision-based intelligent surveillance system, capable of analyzing video streams. These streams are continuously monitored in specific situations for several days (even weeks), learning to characterize the actions taking place there. This system also infers whether events present a threat that should be signalled to a human operator. However, the implementation of advanced computer vision algorithms on embedded systems with battery life is a non-trivial task as such platforms have limited computing power and memory. The concept of the ABORAT system is to apply intelligent vision algorithms on images acquired at the system’s edge (the camera), thus reducing the workload of the processor at the monitoring station and the network traffic for transferring high resolution images to the monitoring station.

we present a smart camera system (ABORAT), with an intelligent processing architecture (ABORGuard) Video Processing Unit (VPU) placed next to an IP camera for processing real-time images, which will then generate and send alerts to the control/monitoring station (ABORGuard Server). The ABORAT system detects and tracks moving objects such as persons/automobiles, collects their trajectories and classifies the behaviour using an autonomous behavioural identifier.

Advantage:

Ø  live remote video monitoring

Ø  live remote viewing from any PC

Ø  Monitor Alarms

Ø  cell phone SMS alerts

Ø  IP security camera use less equipment

Ø  If any person crosses in front of your camera, the software will alert you.

Ø  Store snapshot from web cam.

Ø  Remote access can be easier

Requirements:

Hardware Requirement:-

                       Hard Disk                     -           20 GB

                       Monitor                         -           15’ Color with VGI card support

                       RAM                            -           Minimum 256 MB

                       Processor                      -           Pentium III and Above (or) Equivalent

                       Processor speed            -           Minimum 500 MHz

                                                            -           IP Camera

            Software Requirement:-

                       Operating System         -           Windows XP Professional

                       Platform                        -           Visual Studio .Net 2008

                       Database                       -           SQL Server 2005

                       Languages                    -           c#.net