This is the first issue of a quarterly column I will be authoring for Security Technology Executive dealing with new technologies, technology trends and other issues as they relate to physical security technology. For my first column, I’d like to address two areas in perimeter surveillance technology, one new and one seemingly mature — video analytics and fiber optic sensing.
There is arguably no technology in the security industry that has attracted more hype, more companies and more promise in the recent past than video analytics. While some may point to spot successes, overall revenues have not been commensurate with the promise. I believe that, much like the ballyhooed concept of convergence, its day will eventually come. One reason for the underwhelming market acceptance is the frequent failure to conjure effective algorithms to deal with the entire needs of the application. These needs usually go beyond the obvious. Also, while some analytic algorithms may in fact address one or two applications very well, a limited overall scope has been a barrier to leveraging that into a viable, profitable business.
When it comes specifically to outdoor perimeter sensing, there has been a general failure to assemble a group of interrelated video-based algorithms to deal with the multi-faceted challenge posed by the outdoor environment. Outdoor video detection must overcome a variety of issues, mostly caused by Mother Nature — such as wind, rain, vibration, background motion and scene clutter. A shaking camera creates constant motion in the video scene, greatly complicating detection, analysis and video compression, which relies partly on processing the changes in that scene. Similarly, wave motion, blowing leaves and trees swaying in the wind create a background of constant motion, complicating the job of the video signal processors and the compression engine. Fortunately, systems are now coming to market that have brought the impressive processing power of today’s digital signal processors (DSPs) to bear on these technical issues. This applied DSP technology, combined with effective detection components and processing, geo-registration, improved video compression, bandwidth management and auto tracking forms the basis for some very exciting and effective products.
Long range infrared or visible wavelength detection, coupled with image discrimination requiring only a few pixels on an intruder, provides significant range extension over prior-generation systems. With the resulting requirement for fewer detection cameras, net savings is realized due to lower installation and maintenance costs. Geo-registration enables accurate display on easily available, very detailed maps (thanks, Google Earth!) and integration with PTZ target tracking cameras. Situational Awareness (SA) systems may thus be enhanced as these outdoor detection and tracking mechanisms are added to their other capabilities. As the emerging video standards efforts bear fruit, integration with a broad array of cameras, video management systems and SA systems should become easier and more widespread.
While fiber optics still has the aura of high technology, it has, in fact, been employed in perimeter sensing for years. Taking advantage of the micro-bending properties of optical fiber, techniques similar to those used in an Optical Time Domain Reflectometer (OTDR) — pinpointing a change in optical loss within feet — have been effective in detecting intrusion events. In certain cases, fibers have been “woven” into the fence. In others, “sensing” fibers have been deployed in limited areas.
Limitations in the technology have included installation costs, limited distance, susceptibility to false alarms under certain conditions, and the need for special fibers and dedicated fiber optic cables. New systems, which take advantage of telecommunications-grade components and improved signal processing, use advanced techniques to eliminate many of these issues.
While direct burial or fence-secured systems are still labor intensive, some manufacturers have implemented systems which use standard fiber optic cable, dedicating limited numbers of fibers in the cable to the sensing function. This means that the same cable used for communications — voice, video or data — can also be employed for sensing, simply by allocating individual fiber strands in the cable to that purpose. This can significantly reduce the incremental cost of implementing the sensing function, or it may simply build in future capacity for communications if properly planned. Since unrepeated sensing/transmission distances of 100 km (67 miles) have been reported, the technology is applicable to supervising long stretches of rail bed, highways, utility lines and the like.
Also, if intrusion sensing is not the immediate objective, the same technology can supervise critical fiber optic communications infrastructure, providing early warning for an impending event. Further, the technology lends itself to wide area sensor communication networks, offering the promise of coupling in various types of physical sensors and audio discrimination.
Both of the technologies mentioned above have their preferred applications and, in some cases, may complement each other. It is both interesting and encouraging that, in the most challenging surveillance environments, excellent proven detection tools now exist to provide required levels of performance. These and other perimeter technologies will be on display at ISC-West, and I encourage readers to seek them out, witness a demonstration and challenge manufacturers to prove their capabilities. These technologies have enough promise that your discussions may even lead to new applications!
Ray Coulombe is Principal Consultant for Gilwell Technology Services, providing product and market-related strategic assistance to early stage companies in the physical security and transportation markets. Ray can be reached at firstname.lastname@example.org or through LinkedIn.