Best Perimeter Video Practices
The right technologies and integrator practices help overcome common problems and increase perimeter video systems’ reliability.
Because perimeter video systems are used in vastly different — and much less controlled — environments than what we think of as more traditional surveillance applications, the underlying principles of the two are almost in complete opposition.
“Perimeter protection really is an entirely different animal than standard surveillance,” says Matt Lackrone of Springfield, Mo.-based NetWatch. “Typically, the closer to an object we can get, the better. If we have an option to hang a camera at five feet or 50 feet from a door, we’ll go with five. It’s all about detail, detail, detail. But with perimeter, a lot of times it’s in vast areas where there’s no motion for long periods of time. Do you want detail? Well, how much are you going to get looking at a tree line at a power plant?”
An added complication is that projects requiring perimeter video systems tend to occur irregularly, which can make it difficult to remember the unique factors, considerations, technologies and challenges that go along with these systems.
“Perimeter video is a difficult thing to do. There’s a lot of room for unforeseen issues like weather or even an underground utility that was not marked or was left over from a previous private installation,” says Mark Guzan, security sales manager for MidCo Systems in Burr Ridge, Ill.
Today’s Perimeter Systems
For the most part, the incorporation of video, combined with technological advances, has brought perimeter security a long way from the early days, says Jeremy Bates, general manager and co-owner, Bates Security, Lexington, Ky.
In an effort to improve the accuracy of perimeter applications, Bates says his company won’t even consider installing and/or monitoring a video system at a site with no clearly defined boundaries.
“We’ve learned from experience that the best application is a fenced environment. Monitoring an outdoor space without a fence, where someone can just walk through, causes lots of problems,” he says. “We make sure the customer has a fence or that the area is somehow contained. If not, we don’t offer monitored video.”
That type of combined approach seems to be the most common application for perimeter security, and video plays a large role in that, says Jeff Davis, senior vice president — Western Sector, Kratos Public Safety and Security Solutions, based in San Diego.
“These systems are most often deployed with a layered approach, which may include the integration of fence detection, intelligent video and analytics, and access control systems,” he says.
Customer size and requirements, along with budget, are the real determining factors on what type of approach to employ with a perimeter video system, Guzan says.
“For the smaller-scale customers, there’s often no fence. But for the larger customer, like a natural gas pipe station, they’re using fence protection plus video analytics if a multi-layered approach is needed and if there’s a budget for it,” he says. “Not everyone needs a fence and video, and sometimes they’re restricted on the type of fence they can use.”
Something’s Out There
First and foremost, perimeter security is about detection. But simply detecting that something entered a specific area — whether fenced or not — isn’t enough. The second part of the equation is identification, which is where perimeter video comes in. Now, we’re not necessarily talking about the level of identification you’d want from traditional surveillance, although that can be accomplished in a perimeter application. In these cases, it’s more often than not detecting whether whatever has entered the area is indeed a human.
“The ideal concept is to create an overall visual perspective of the perimeter with the ability to immediately focus on a specific point in the perimeter as needed,” Davis says. “The goal is to efficiently and effectively identify the threat or risk and react with appropriate operational response.”
Improvements in video analytic technologies have helped make perimeter video much more reliable and less prone to problems.
“As technology has progressed, we’re starting to utilize more video analytics, especially for applications where we had challenges with false alarms,” Bates says.
When combined with thermal cameras, analytics become that much more powerful, Lackrone says. With motion detection, the camera is concerned with how many pixels in the scene change, and with thermal cameras those pixel changes are much more prevalent despite using VGA-type resolution. Never mind the fact that thermal cameras can detect intruders over longer distances. Or that they can see in the dark.
“People do things at night, and even as good as cameras are with wide dynamic range, you still have more instances of false positives and image quality over distance,” Lackrone says. “Thermal cameras get information to where you’re able to get out there, find a person, stop a theft and get back to normalcy.”
Ashland Integration, based in San Bernardino, Calif., installs perimeter video systems almost exclusively in remote desert environments. Because of those factors, the company uses all of the above, plus fence detection technologies, says Laura Terman, Ashland’s director of operations.
“A typical system design we use quite often is fiber on the fence coupled with thermal and analytics. So we have sort of a 1-2-3 process where the fiber on the fence detects the possibility of intrusion. The camera then swings to that zone. The analytics and the thermal give a real clear, crisp image, and then that’s transported to a monitoring station or monitoring company,” she says. “And because the analytics and the thermal worked so well together, it gives the humans involved in the process a real clear idea of whether or not there has truly been an intrusion about to happen or whether it’s a coyote or a jackrabbit.”
The next logical step is determining whether and how to act when an object has been detected and identified as a human being, Davis says.
“As with any electronic security system deployment, the key element to the system is a well-defined standard operating procedure. Electronics only work as well as the operators controlling and reacting to them,” he says.
Different customers are going to have different approaches that work for them, so there’s no standard answer to the question of response, Lackrone says.
“There’s something to be said for how the end user operates the camera with an asset in the middle of nowhere. They’ll know there’s been activity there, but whether they know it’s a valid breach or a false positive depends on how the entity manages cameras. Is someone monitoring all the time or should someone be just getting a text? There’s variance on how to use the system. If a guy who owns a manufacturing facility gets a text at three in the morning, he’s less likely to run out and check that out than a guard who’s manning a utility.”
The right response can and does make a big difference, as UCIT sees on almost a daily basis.
“We have all kinds of perimeter video solutions, some with fences and some without; regardless, the main purpose is to detect a breach in real-time in order to avoid any losses. When there is a breach we work with local law enforcement; we currently average more than four arrests per week. The video is then also used as evidence once the individuals are caught,” says Sidney Sommer, managing director of the Mississauga, Ont.-based UCIT (pronounced “you see it”).
If you ask someone to name the challenges involved with installing a system — whether it includes video or not — in a perimeter application, there’s a good chance they’ll say things like weather, animals or other environmental factors causing false positives. But there are several other, less well-known challenges that could arise on just about any project and raise the blood pressure of the installer.
One of those is camera placement, which when combined with tunnel vision can completely derail an installation.
“Sometimes you can’t put cameras where you want them,” Guzan says. “Maybe you can’t because it’s landlocked by concrete or there are utilities in the way. In that case, you have to just find the best alternative and use that.”
In some areas, connectivity can cause headaches, if not ulcers. Until recently, that’s been one of the biggest issues NetWatch has faced. Making it all the more frustrating, Lackrone says, has been watching competitors in not-so-distant cities have access to faster wireless networks while his company’s service area lagged behind.
“In the Springfield market, it’s only been in the last few months that we’ve really had reliable LTE available, while the Kansas City and St. Louis markets have probably had LTE for a couple years,” he says. “With a camera and an LTE modem, you can hook them up and get reasonable service. But with a single camera and a 3G modem, the communication is not that good — especially for people who need perimeter security.”
Those customers include owners of high-end livestock, some of which go for around $10,000 a head for a bull, Lackrone says. To protect that kind of investment, they need a perimeter video system, but the communication structure isn’t always there. However, he expects that to change in the near future.
“People who need perimeter security have something that many times is far away from the infrastructure to pump back to an Internet connection somewhere. That’s always going to be a challenge, but with LTE modems, we’re going to have more success,” he says.
Lighting is a big problem for UCIT, particularly with the large number of solar fields the company works with.
“Some of these fields are 200 acres and located in the middle of nowhere. Lighting it up at night would not be cost-effective and would be a distraction for people living in the area,” Sommer says. “We designed a solution using Axis thermal cameras together with some day/night PTZ cameras. These customers in the past would have used an on-site security guard, which would have been ineffective in covering a 200-acre field. Our solution is less expensive and alerts our monitoring center instantly when somebody walks up to the fence day or night. Our customer also uses the cameras as a project management tool, for example to check if there is snow on the panels or if they are moving correctly.”
For more insight into the types of challenges integrators face with perimeter video, see “Clearing the Hurdles” on page 86.
The best way to deal with issues — both anticipated and unexpected — that are bound to arise with any perimeter video installation, is to completely understand them ahead of time.
“We partner with our customers to understand their needs. Sometimes we’ll be out with a customer and see work being done at their facility, maybe blacktop or a new parking lot,” Guzan says. “So we’ll tell them, ‘Now is a great time to put a camera on a pole.’ If they don’t want to right now, they can plan ahead by putting the infrastructure in place, like an extra conduit to allow them to do it in the future. That way, when they expand, they’ve already overcome one of the biggest challenges.”
It’s also imperative that the customer accepts his or her responsibility in making sure the system continues to function properly, which may have nothing to do with technology.
“Weeds growing up create false positives. Parked vehicles cause motion problems. So they have to commit to maintaining a good, proper perimeter fence,” Bates says. “They have to make sure the things that cause false activations get addressed. Otherwise, they’re training operators to think that all alarms from that site are false.”
The best thing to remember about perimeter video systems may be to remember the old adage about forcing a square peg into a round hole.
“The system has to be proper for the environment, and not every outdoor location is a good environment,” Bates says.
Clearing the Hurdles
Any number of factors could potentially delay or scrap a perimeter video project altogether. Being prepared for them helps a lot, so here are a few things integrators have encountered in their installations.
“Occasionally there are certain expectations of the system to capture very specific activity in accordance with regulatory compliances. Another challenge as it relates to the installation is the requirement to utilize proper product selection for the application — lenses, camera technology and quality. Many times a solution provider is too committed to a particular brand of product as opposed to providing the proper solution for the application. An open mind and a working knowledge of solutions and technologies available in the industry are imperative.” — Jeff Davis, Kratos Public Safety and Security Solutions
“In general, weather is the biggest challenge when installing an outdoor video system, especially one that gets monitored. It’s very difficult to program the cameras to send alarms when there is a true event and not when it’s windy, rainy or snowing. The added challenges for some of the perimeter applications are the actual distances of the perimeters, which can be kilometers, and lack of power along the way.” — Sidney Sommer, UCIT
“Cabling distance is a big one. When you’re using IP video camera out at a perimeter, you most likely have to utilize fiber and transmit that to copper. Along the same line are power concerns, which depend on how far you’re getting cable or power. You either have to trench or directionally bore. In our climate, that goes along with weather conditions. We have to start pretty soon in the fall or we’ll miss the window before the ground freezes.” — Mark Guzan, MidCo Systems
“Detail. The sensitivity has to be so very high that to alert on motion, you can get a lot of false positives. If the sensitivity is too high, when a bird flies by at 50 feet, you could get a false positive. If a man is one-inch tall on a screen, and that’s what you want to detect, it’s not hard to set the system properly.” — Matt Lackrone, NetWatch