Video Verification Is Not a Silver Bullet

Oct. 27, 2008
Why cameras don’t get false alarms out of the picture
Untitled Document I am a big fan of technology. I really am. Point out a problem, and my mind jumps to an electronic solution. I guess that’s why it upsets me so when a new product concept like video verification, in spite of many promises, fails to solve one of the industry’s most pressing problems: false alarm avoidance.

Most of the public—and, for that matter, many security directors and other users of security technology—have a tough time relating to the false alarm problem. We all know it’s there, but most of us underestimate it. That’s part of the reason the recent rise in verified response laws nationwide has been so controversial.

Intrusion? Prove It.
Verified response legislation allows police departments to withhold response to alarms that are not verified as valid. Dallas joined the ranks of cities with such legislation in December of 2005. In defending the move, they stated that of the 62,000 burglar alarm calls the police department received last year, 60,100—that’s 97.2%—were false alarms. While many would criticize the solution Dallas chose, it’s not hard to understand why they felt compelled to do something.

While the Dallas legislation calls for verification by an eyewitness, “the city attorney has not decided whether to accept video verification as that eyewitness,” said Donna Hernandez, Dallas Police spokeswoman. And Dallas is not alone in driving a solution. Approximately 24 other communities have enacted similar legislation. As the budget pinch grows tighter for local police departments, they are becoming more vocal about the workload the alarm industry places on them and their perception of its lack of progress on the issue.

Surprising Improvement
Of course, the reality is, there has been significant progress. The industry has developed new sensors that are less prone to false alarms. They have upgraded panels to the new CP-01 standard to ensure the defaults and timing are set to minimize alarms. Many central stations have instituted enhanced call verification, which doubles the number of calls to the end user before the police are dispatched. “We are seeing 35% to 50% reduction in dispatches with a second call,” said Stan Martin, executive director of the Security Industry Alarm Coalition. Still, even though the number of alarms per system has dropped dramatically, dealers are installing so many new systems that the load on police is still going up.

It was in the middle of the 2004 proposal from Los Angeles to go to verified response that many in the industry started to look seriously at remote video as a potential solution. The challenge, it seemed, was making the equipment and transmission costs low enough. Since that time, many companies have developed systems that are well integrated with existing alarm systems and that lower the costs significantly. What is now becoming clear is that costs are not the only issue.

If the response to an alarm is to send someone to look around inside the building, why not use video to look around before we dispatch the police? Like a lot of things in life, the devil is in the details.

The Problem of Coverage
First, it turns out that “looking around” with video is not that easy. Coverage is a problem. How much is enough? Installing multiple cameras and PTZ domes will get you a picture of most everything, but the costs for this would climb so high that many alarm users wouldn’t be able to afford it. Multiple fixed cameras might be cheaper, but how many does it take to cover a building?

We often hear that 100% coverage is not necessary; you just need to cover the obvious points of egress. Well, to see why that is not true, let’s put ourselves in the position of the central station operator receiving an alarm.

While the key objective is to avoid burglaries, when it comes to the everyday world of false alarms, the objective is to avoid an unnecessary police dispatch. To meet that objective, the central station operator must prove to himself or herself that either there is no one on the premises or anyone on the premises during an alarm belongs there. Using video to prove either of these is difficult.
Verifying that no one is on the premises requires enough coverage to feel comfortable that someone is not hiding out of view. But few facilities have coverage even close to that complete. Picture a school with hundreds of classrooms. “If someone put video verification as a mandatory part of an ordinance, it would be extremely costly for the average commercial office building,” said Martin. How would you economically provide enough coverage to be able to say that no one is in the building in spite of an alarm event?

Just as important, what central station operator would take the risk of not calling the police if the video shows nothing? “You have to make a judgment call. Is that person supposed to be there or not? That’s a problem. Central stations are used to having an alarm on zone 3 and calling the police,” said Lou Fiore, president of LT Fiore Inc. and former president of the Central Station Alarm Association.

Who Is That Masked Man?
Let’s go back to our theoretical central station operator. Let’s say some video comes in from a small retail establishment. There seems to be a nervous man with a gun going through drawers. Is he a burglar or the owner of a shop in a bad neighborhood who forgot his wallet?

We know that most false alarms come from user error. Does a picture of a possible member of the cleaning crew at the keypad really help? Not according to SIAC’s Martin. “Now you have a picture of the false alarm,” he said. Video helps to confirm someone is present, but voice communication is still required to confirm that person is authorized to be there. One can argue that if you still need voice, the video doesn’t help all that much.

One-Two Punch: Privacy and Cost
While mostly a residential issue, privacy is the key reason video verification will never be a universal solution to community false alarms. Most people will simply not be comfortable putting cameras in their homes. While some have said that the cameras need only be near the doors and hallways, think about where most people keep their valuables (master bedroom) and then ponder the wisdom of not needing a camera there. While it seems that the privacy issue does not affect the commercial market, it does in a very real way. Cities are not likely to mandate a solution unless it works in both markets. Instead, they opt for solutions like private responders.
Cost is also still an issue. These systems start at the high end with many thousands of dollars of conventional cameras and recording gear, and go all the way down to board-level cameras that connect to add-on controllers in the intrusion panel. The low-end systems offer “still views” of key areas before and after an event, and are about as cheap as the technology will allow in the near future. Even at that, they generally double the cost of a small intrusion system.

“The problem is, the solutions are too expensive. A $300 to $800 intrusion system is somehow not good enough unless you spend another $1000 on video,” said Gordon Hope, vice president of marketing for Honeywell. “In and of itself, video does not fix this problem.”

So When Does Video Help?
For alarm verification, video can help particularly in small facilities like retail or tenant offices where 100% coverage can be achieved with relatively few cameras. “We have a lot of high-end customers in Manhattan. They see the value. It also protects my response people by letting them know they are walking into a potential problem,” said Craig Decher, vice president of DGA Security Systems in New York.

Beyond alarm verification, of course, remote video does have a significant number of potential uses. At the upper end of the market, high-end cameras and DVRs allow full site monitoring off-premises. Cost savings through centralizing or outsourcing that virtual guard tour function are real.

Management can gain real value by monitoring their smaller businesses from off-site to look for point-of-sale events and operational audits, or even to gauge the level of business. Remote video also works to monitor places where humans should not be, such as outdoor perimeters at airports.

How Do We Solve False Alarms?

Could video be made to solve the problem? There are many cases where it helps today. But to get to a universal solution that could be mandated, we have a long way to go. Costs would need to go much lower. “The lower the price, the more opportunity there is for video verification,” said Martin.

Some mechanism would have to be added to authenticate that a person really is one of the authorized site occupants. While a facial recognition solution would be ideal, any assistance for the operator would help. Solving the privacy problem, unfortunately, will be the most difficult of all. The only way to gain almost universal trust would be if the video never left the premises. The system will have to get smart enough to make those judgments on its own.

In my business of helping manufacturers with product strategy, I often see companies get so enamored with a technology that they forget the purpose is to solve their customers’ problems. It’s time for the industry to take another hard look at why we struggle with this issue. We know most of the false alarms come from user error. Any time users make billions of mistakes a year, the problem is not with the users; it is with the design of the system. The answers are not simple, but neither is convincing people to pay a monitoring fee without police response.

Rich Anderson is president of Phare Consulting, a firm providing technology and growth strategies for the security industry. A 25-year veteran of high tech electronics, Mr. Anderson previously served as the VP of marketing for GE Security and the VP of engineering for CASI-RUSCO. He can be reached at [email protected].