ERCIST Intelligent Video Surveillance System

Preface

The latest generation of computer vision technology is revolutionizing concepts, applications, and products in video surveillance and CCTV. This is of prime relevance to security for large outdoor facilities such as commercial airfields,refineries, power plants, and office/industrial campuses. Most airfields, for example, have open (unfenced) perimeters, high volume heterogeneous traffic,are easily accessed on foot or by water, and exist in areas where regulations providing a safety buffer are difficult to legislate or enforce.And all airfields require 24x7 outdoor monitoring–snow,fog,rain or shine. Likewise, most high-value facilities appealing to criminals and terrorists are in close proximity to public areas (roads, residences, city, etc.).
The appeal of automated real-time surveillance is obvious–maximizing efficiency and effectiveness of security personnel and resources while increasing the probability of preventing a serious security breach. Computer vision based solutions have the potential for very discriminating detection and very low false alarms. The bottom line is that applied computer vision has the potential for the greatest return on investment(ROI), both short-term and long-term.

ERCIST integrates the computer vision technology and more advanced DSP chip technology, proposes an ERCIST Intelligent Video Surveillance System solution EIVS. EIVS solution watches video streams and extracts descriptions of all relevant objects. It employs sophisticated algorithms for detection and tracking of all relevant objects in the camera's view. It also contains algorithms for classification of objects into specific types and illustrates an image from ERCIST's application not only for antiterrorism but also the criminal detection.

EIVS Abstract

Several years prior to the successes of applied computer vision, manufacturers of Digital Video Recorders (DVRs) and other video processing solutions began a dubious flirtation with a technology called Video Motion Detection (VMD). This technology analyses video imagery and determines where there is motion in the scene. The theory being that anything that moves is something interesting to users. What quickly emerged was the result that there is a lot of motion in the world that is highly annoying and something more was needed. In a situation such as detection of watercraft, the entire image is moving and VMD gets very confused. A more intelligent AVS system can accurately extract the relevant information from generic motion.

The solution to this problem is automated video surveillance (AVS). That is, computer software that watches video streams to determine activities, events or behaviors that might be considered suspicious and provide an appropriate response when such actions occur. The key technology is called Computer Vision. This is a somewhat obscure branch of mainstream artificial intelligence research involving teaching machines to understand what they “see” through a camera. Traditionally, computer vision has had limited success in real-world commercial applications, but recent advances in technology and computational power along with a move of key talent from academia into industry have allowed computer vision to come out of the lab and into commercial video surveillance products.

Comparing with current DVR and VDM technologies,EIVS is a powerful real-time intelligent video surveillance software that, according to User-defined rules,

1. Detect and identify
2. Object-of-Interest:people,vehicle, watercraft, etc
3. Event-of-Interest:crossing a video tripwire,or entering a specified Area-of-Interest,or appearing in the camera view,etc.
4. Determine if Object-of-Interest/Event-of-Interest are security risks to User.
5. Send alerts to security personnel via pre-defined channels:Alert Console,telephone,mobile phone,pager,PDA,email,etc.

The standards-based hardware design of EIVS and its robust API frameworks enable plug-and-play with most surveillance devices,or existing infrastructure. With such distribution of intelligence across the security surveillance infrastructure,the overall system complexity and cost of ownership is greatly reduced; and the ability to selectively output only alerts and other actionable information,means the network bandwidth is optimized.

System Architecture

ERCIST Intelligent Video Surveillance System delivers the functionality as an embedded application running on a high-performance digital signal processor (DSP). It transforms the data into highly intelligent devices with built-in intelligent video analytics at the shortest possible time, and minimal costs.

 

Solutions

It is as sample as cabling a wire from the camera to the EIVS that will automatically identify the video data based on the user-defined rules, they proactively analyze video feeds, and produce alerts and other actionable information at real-time. Once the Event-of-Interest is security risks to User, the proper action will be taken place by security personnel.

Components

Intelligent Video Process Unit (IVPU)
IVPU is a core chip of ERCIST Intelligent Video Surveillance.It accesses the video signal from the camera and separates the objective from background information for identifying, analyzing, and tracking. It compares the activities of objective with the setup rules. It triggers the alert, record the event immediately through the network if an abnormal condition occurs.?

Server
It records the detail data that the alert is triggered by the EVIS.

Strategic Management (SM)
It includes a Rule Management Tool (RMT) that enables the User to easily define rules for any camera view. Rules instruct the Video Processing Unit (VPU) what security-related events to monitor on surveillance cameras. A rule consists of an event,a schedule,and one or more alerts. If an event occurs during the period of time specified in the schedule, EIVS will generate an alert.

Alert Console (AC)

It is installed in the PC and the alert information will be displayed on the screen once EIVS receives the alert signal from the server. It shows the time of the alert occurred, video channel, the ID of the EIVS and two frames of video pictures that indicates the event. The alert information have guard's wits about event therefore improve the effective and efficiency of security of management.

Function

EIVS can distinguish the Object-of-Interest in people, vehicle, watercraft, etc via the video data from camera. It can filter out the distortion from natural phenomenon, such as the distortion caused by water glisten, tidewater, wind caused. Therefore, it reduces the negative false. There is a true abnormal event existing when EIVS delivers a signal, then the proper action should be take place. The different video security rule should be set based on the time and location, for instance Bank, Hotel, on/off peak time, etc

	Identify and sort: It monitors the motional objective(s), such as person, vehicle and others via the camera from the screen. The system will track, sort and analyze the objective to identify if it is any unsecured situation there. Polygon Area: It monitors the person and vehicle from a special area, such as deferent area, boatyard, train station, backyard of shopping mall, etc. A single virtual tripwire and Multi-tripwire It is used to watch out the person or vehicle that crosses the trip wire(s). The example for a single tripwire for example security line, one way entering, railroad fence, and so on. The multi-tripwire can be applied in some special event such as violated left turn in the traffic road, speeding vehicle, etc.
	Loitering System will generate an alert? to the security personnel when person(s) or vehicle(s) linger the security area for a while without any reason.
	Stealing Monitoring valuable materials, it can be applied for national or local exhibition center, unmanned cell site, Equipment,? Manufactory, Armory, etc.
	Left behind Monitors if the object was left in some public areas that may cause homeland security issue.

specification