Three views of video surveillance
Depending on who you talk with, the video surveillance field is either leapfrogging forward, bogged down in unfulfilled technology promises, or utterly confused. After four days in Las Vegas, and dozens of conversations on this subject, I’ve concluded that all three views are somewhat accurate.
On the positive side, the video surveillance field has found a new buzz word -- H.264-- to get its collective heartbeat racing. H.264 is the name given to the latest generation of video compression standards. This particular standard was originally established back in 2003, and initially taken up by the television broadcasting industry, but has recently found its way into the security world.
At ISC West, Axis Communications, one of the leaders in IP-based video networking technologies, made a splash by announcing the introduction of its new ARTPEC-3 chip, which it says uses H.264-compliant technology to deliver a high-quality 2.5 mexapixel image at 30 frames per second -- about three times the performance of any earlier compression chip. That sounds good because better compression in a camera, video encoder or video management software can speed video transmissions, gobble up less bandwidth, deliver a clearer image, require less storage space and, ultimately, save the operator a lot of money.
P-based cameras, like those offered by Axis, currently represent only about 15 percent of the total video surveillance market, according to company execs, but as compression technologies improve, more and more cameras are expected to migrate to IP-based technologies.
"Fifty percent of the market will be based on H.264 products within two years," predicted Ray Mauritsson, the president and CEO of Axis.
Of course, not all observers were as enthusiastic about the arrival of this cutting-edge compression technology. Some realists, such as Jean-Pierre Forest, director of security solutions for Aviglon, of Vancouver, Canada, told me that high-quality images delivered at faster and faster frame rates per second were not the be-all and end-all of video surveillance. Forest, who comes from a video forensics background with the Canadian military and several Canadian police forces, believes that it is better to have a single, sharp, high-resolution video image that enables an investigator to make a positive identification of a suspect, than hundreds or thousands of images that are far less clear.
On another front, a second set of newly-fashionable terms -- "open source," or "open architecture," or simply "open" -- seems to have fully blossomed at ISC West. In earlier years, only the avant garde technology-driven companies, staffed by refugees from Silicon Valley, boasted that they embraced the philosophy of open architecture. Last week, however, it seemed that every Tom, Dick and Harry at every booth I visited had quaffed the Kool Aid and now boasted of their adherence to an "open" approach in the video surveillance marketplace.
What, precisely, does that mean? It’s tough to say. Some companies use the phrase "open" to suggest that they’re willing to partner with other firms in the surveillance niche to pull together a hybrid "best-of-breed solution" that will offer the customer exactly the set of components it wants. Other companies use the phrase in a more technical sense to suggest that their component will have an API (that stands for Application Programming Interface), which will allow their system’s software to interface, or "talk," with another company’s software. In either case, it has now become de rigueur in the surveillance field to profess a willingness to cooperate with other companies -- including some competitors -- for the greater good of one’s customers.
Some observers aren’t buying it. "There’s ‘open’ and there’s ‘open’," said Mariann McDonagh, senior vice president of global marketing at Verint Systems Inc. McDonagh explained that the video surveillance niche began with separate technologies existing in separate "silos," which typically were unable to talk with one another. Some companies might claim to be offering an "open" solution -- "your hook connects to my hook" -- but how truly open is that solution if all communications pass in only one direction. (For example, an access control system can signal the video surveillance management software that a door has suddenly been opened, but the video software cannot send a command back to the access controller to lock all doors.)
In McDonagh’s lexicon, true openness would require bi-directional communications between cooperating systems, a situation she rarely sees.
Similarly, Robert Beliles, a formerCisco Systems Corp. employee who took a post as vice president of business development at Hirsch Electronicsearlier this year, is highly skeptical of this new-found fondness for openness. "There is no vendor on the planet in their right mind which is totally open," said Beliles. To be fully open would mean that a company would allow outsiders access to every line of its source codes, a policy that would be risky, time-consuming and expensive. "The reality is that ‘open’ companies are open only at certain points," he added.
There may be advances in compression technology to crow about, but the folks offering "video analytics" -- sometimes called "image analytics" or "intelligent video" -- now seem to be on the defensive. For the past two or three years, these video analytic companies have been touting their software packages as capable of solving a myriad of security problems -- such as spotting left luggage, reading license plates and identifying people walking in the "wrong" direction -- but too often the reality has fallen short of the promises. Some have blamed unpredictable lighting conditions. Others have pointed to the vagaries of weather. While others have fingered the lack of sufficient processing power to crunch the massive quantity of available pixels. It will be interesting to see whether improvements in video compression technology can lead to better performance by the industry’s image analytic algorithms.
It seems to me that many video surveillance suppliers -- along with vendors throughout the security industry -- have finally begun to recognize that their "solutions" are often too complex for their government customers to understand. Buyers are bewildered by the array of components that need to be integrated, the slew of standards that need to be complied with, and the utter impossibility of pinning the blame for a system’s failure on any one vendor.
"Our customers are confused by all these choices," Mark Provinsal, vice president of strategic marketing for Dedicated Micros, told me. "Often, the product doesn’t do what they thought it would do."
It used to be that exhibitors at ISC West competed with one another by offering the most bells and whistles, or the "end-to-end solutions" with the greatest number of mind-boggling new features. At the ISC West show last week, however, the opposite theme seemed to be emerging: Keep it simple, stupid. Less is more.
As Verint puts it in its latest marketing campaign: "IP Video Made Easy"
- Add your comment
- trackback url: http://www.gsnmagazine.com/cms/trackback/651-1
