How to Manage Surveillance Video
- 14 May, 2009 10:55
Video management software (VMS) allows you to record and view live video from multiple surveillance cameras--either IP-based or analog cameras with an encoder--monitor alarms, control cameras and retrieve recordings from an archive. Because they are IP-based, VMS systems are more expandable and flexible than DVR-based systems, and employees can control the software from anywhere on the network. Surveillance and security teams can use the software for live monitoring, as well as investigative and forensic purposes, using archived footage.
Users have three form factors from which to choose for managing IP video: software-only, hardware/software appliances (sometimes referred to as network video recorders, or NVRs) or a hybrid DVR, which is a DVR with additional software to manage IP equipment.
Because of the economic downturn, the VMS market will see slower growth in 2009 than in previous years, with a forecast of 29 percent versus more than 40 percent, according to IMS Research.
Evaluation Criteria for Video Management
VMS systems range from the basic to the sophisticated, with major differences including reliability features and number of cameras and locations supported. Here is a sampling of features to consider:
* Specific options for different verticals, including retail, banking, transportation, etc.;
* Video analytics, such as license-plate or facial recognition;
* Integration with third-party systems, such as access control, building automation, alarm management, video analytics and more;
* Motion detection;
* Customizable, resizable viewing panes;
* User interface features that include hot-spot windows, color-indicated activity, instant replay, quick switching between cameras, etc.;
* "Privacy zones" to protect sensitive areas from being monitored;
* Creation of customized rules. For instance, if a particular door opens, the camera begins recording and even activates an alarm or sends an alert;
* Camera control (pan, tilt, zoom);
* View multiple video channels at once;
* Multichannel playback, which allows users to play recorded video from several cameras simultaneously--useful if tracking a suspect through hallways;
* Multiple search devices, including fast-forward, reverse, thumb-nail view, time line bars, bookmarking, etc.;
* Secure export of material evidence;
* Fail-over capability that enables continued recording if the primary server goes down.
Software-Only Versus Appliance
Users can choose between software that they load, configure and manage on a server of their choosing or a hardware appliance that's preloaded with software. The benefits of appliances are reduced setup and installation complexity, while disadvantages are less flexibility, fewer customization options and more difficult integration with third-party systems. According to Simon Harris, senior research director at IMS Research, more advanced users will typically opt for software-only solutions, while those that aren't comfortable doing setup and configuration will choose an appliance, or what IMS calls a proprietary system.
"If you have only a small number of cameras and don't intend to integrate with your access control or building management system, that lends itself to proprietary systems," Harris says. "As they get bigger and more complex, that's when they go for open platform."
This is what Jeff Hinckley, a systems integrator at Norris, ran into when he designed and implemented a video surveillance network for the cities of Lewiston and Auburn in Maine. The schools were working in conjunction with Lewiston-Auburn 911, an organization created by Auburn and Lewiston to provide dispatch and radio communications to first responders. The school system had obtained federal funding for a surveillance system to help with increasing criminal mischief.
The surveillance system was designed to work with a large-scale, 4.9 GHz wireless mesh network designed by Norris and deployed throughout both cities. The network was intended for emergency access to municipal security systems as well as the deployment of remote cameras. Early in the endeavor, some of the Auburn schools were equipped with analog cameras and DVRs from Pelco, as well as some IP cameras managed by a Pelco video management appliance. But as the surveillance system was expanded, it became increasingly desirable to use high-resolution megapixel cameras in some areas and to use a more unified, single-application approach to accessing both video and other security-based services.
Using encoders, Norris tied the analog cameras into the schools' fiber-based network, added megapixel cameras and replaced the Pelco VMS with software from Exacq Technologies. Exacq's system, which runs on a Windows or Linux server, is able to support all the cameras and can integrate with access control systems.
VMS Versus Hybrid DVR
Some DVR vendors have begun selling software that enables their DVRs to support both IP cameras and directly connected analog cameras. VMS-based systems can also support analog cameras, but they require the use of an encoder to translate the signal to digital. With hybrid DVRs, both types of cameras are supported directly.
The hybrid option will be particularly attractive to companies during the economic downturn, when many end users will be motivated to make modest, incremental upgrades to IP, while staying with their existing DVR providers, says John Honovich, founder of IP Video Market Info, a video surveillance information portal. "In the past, if you wanted to add megapixel or other IP cameras to your surveillance system, you were forced to go to an IP-based VMS solution," he says. "That has become much more complicated now that DVR vendors are rolling out increased IP support."
DVR vendors still don't support the breadth of cameras that VMS-based systems do, Honovich says. They might support two to five manufacturers versus leading VMS vendors, which support up to 50. "The DVR folks have a ways to go to catch up," he says. At the same time, over the next 18 months, the distinction between these manufacturers will largely disappear, Honovich says. He sees DVR companies broadening their IP camera support and selling software-only systems, either independently or through acquisitions, while VMS vendors will offer DVR/NVR appliances to appeal to organizations with smaller camera deployments.
Already, he says, hybrid DVR vendors offer enterprise-level functionality, such as centralized management and third-party system support that, in some cases, is superior to VMS vendors, especially in the area of access control. "IP players will say they're more open, but I'd ask them to prove that," he says.
VMS Dos and Don'ts
DO investigate the level of support for third-party systems. VMS systems can integrate with a wide range of third-party systems, including access control, video analytics, building automation and alarm management systems. Companies should investigate not just whether the VMS integrates with the systems it needs, but also the level of integration, which varies widely, says Brian Carle, product manager at Salient Systems, in a blog entry on the IP Video Market portal.
At a basic level, the VMS can receive and act on alarm events from the third-party system, he says. For example, when a person enters a building, an access control system could trigger the video management system to verify that the image of the person captured from video matches the ID card/system. At a more sophisticated level, he says, you'd be able to configure the third-party system through the VMS interface.
DON'T get stuck on a particular vendor until you know which cameras they support. Whereas DVRs support almost any analog camera, such is not the case with IP video software, which needs specific drivers for each camera type. "Some support only one brand, while others support 500," Honovich says. "You can decide you really want to use a particular IP camera but then realize it's only supported by five software vendors in the market."
DON'T forget user authentication and authorization. A big benefit of VMS systems is that you can centrally manage an unlimited number of devices. But you also need to consider how you're going to centrally manage the users accessing the system, especially if they're geographically dispersed. One way is to ensure the system integrates with the directory services you're already using, such as Microsoft Active Directory Support. "If you're already using that for PCs, you can integrate your video surveillance system with that so they're both using the same user name and password," Honovich says. Plus, you can keep logs of video-watching behavior in a database.
In the bad old days, says Honovich, you would have to set up a unique user name and password for every DVR or video management system, which created a nightmare scenario of people using weak default passwords like "admin." "Anyone could get in at any given time," Honovich says. "That's why centralized management is an important element, given the history of poor user access management."
DO look for an intuitive user interface. Security personnel can turn over quickly, so it's important to have a system on which you can train people quickly and easily. Not to mention, many are moving from the world of analog systems, so the transition to a computer interface needs to be considered. "Security can be a fluid profession, so you don't want to invest a lot of training, which makes an intuitive interface paramount," Honovich says.
Often, Carle warns, applications are designed with functions buried under menus, and it takes many mouse clicks to perform a function. "Ease of use and training are a primary concern for organizations that have guards monitoring the video," he says.
The best thing to do is try the product in an existing environment to see what users do and don't like, says Kani Neves, executive director of the Sherwood Valley Gaming Commission. That's what his group did before choosing a system from Genetec for the Black Bart Casino in Northern California. "We visited other facilities and environments to see what we thought would apply to us and not," he says.
DO consider resource-saving features. Some features can help minimize time and staffing levels, including alarm clients, mapping clients and smart searching. Alarm clients display a blank video screen until activity occurs on associated cameras. Only video triggered by motion or alarm will display, "which prevents the operator from being bombarded with potentially irrelevant video information," Carle says. This can cut down on the number of personnel you need. Some alarm clients include a history list of events, so an operator can click on an item in the list and quickly play back video of the event, he says.
Smart search, Carle explains, speeds investigations because you can specify particular areas in the camera's field of view, as well as a specific time frame, and capture only recordings where motion is detected in that area in that time period. Systems differ, he says, in terms of the speed of these searches, depending on whether it records metadata along with the video.
Meanwhile, a mapping feature allows administrators to import an image file and overlay icons representing cameras on the map, Carle explains. "This will show an operator exactly where a camera is in the facility, making it much easier to learn the system and track activity across cameras."
DO understand cost structures. Many vendors calculate cost by charging a certain amount for each video device used with the system, plus an upgrade subscription fee, which entitles the user to download new versions of the software, Carle explains. Some also charge a server fee, either as a site license or for each server on which the software is installed. Per-device costs vary widely, mainly depending on the system's level of sophistication. Very high-end systems can be over $1,000 per camera, while an enterprise-level, scalable system can be US$200 to $500 per camera, Carle says.
DON'T forget to consider storage. Especially with higher resolution cameras, video surveillance can start to take a big bite out of storage. Many vendors offer various techniques to keep that to a minimum. For instance, the Genetec system that Neves selected can change video resolution from 4 CIF to 2 CIF or less. "The software can manipulate what the camera sees and records, how much it records and with how much clarity," he says.
Another way that Neves minimizes storage requirements is through Genetec's motion detection capability, which can be applied to any camera even if the camera itself does not have the ability to sense light. The feature enables the casino to record only when tables are in use, to meet the federal requirement of 24/7 recording for active tables. In all, there are 100 tables, and if just 50 are being used at a time, only the cameras focused on those tables need to record video.
"Anytime a customer or dealer enters the frame, the system automatically starts recording five minutes prior to that," Neves says. "And you can tell it to stop recording once motion has left the frame. It allows us the freedom to utilize cameras when we need them."
Plus, if certain areas of a casino are only accessible at certain times, he can use motion detection on cameras monitoring the hallways and doors in those areas to check for abnormal access. "With 500 to 600 cameras, we don't have the manpower to hire the people it would take to see everything going on," he says. "This enables us to minimize our staffing while increasing our security level."