IFSEC Global

Author Bio ▼

IFSEC Global is the online community for the Security and Fire industry. Our market-leading live events span the globe, connecting buyers and sellers.
March 12, 2009

Sign up to free email newsletters

Download

Mobile access series #1: What you need to know

Video analytics: analytics as a sensor

The video analytics market is growing, the technology is propagating quickly and the number of new providers is increasing by the day. According to an IMS report, The World Market for Video Content Analysis – 2008 Edition, the value of the market is projected at over five billion dollars by 2012.

Video analytics’ reach has widened from the top of the pyramid – military, government, nuclear facilities, airports and seaports – to the mid-range market of industrial facilities, education, logistics, etc. At the beginning, systems were expensive, difficult to install and operate, and costly to maintain. In order to expand into the mid range-market, video analytics vendors improved their products for easier installation and cost effectiveness, without compromising performance.

A product’s ability to permeate from high-end to mid-range markets is often determined by its architecture. In order to penetrate mid-range markets, a different approach is usually required and certain elements need to be taken into consideration, simplicity being the essential common factor.

Systems typically found in high-end markets may not be suited for mid-range markets, mainly due to their higher price and complex integration requirements.

These markets call for systems that are flexible, cost effective and simple to install. Remote setup and maintenance capabilities are also important to alleviate the travel costs of technical engineers usually associated with post installation procedures and maintenance.

The key access point to the mid-range market is through system integrators and installers. They are exposed to the latest security technologies and have access to an important database of existing CCTV installations, potential targets for upgrading to video analytics. A key challenge that system integrators face when dealing with video analytics is the customer’s desire for a system that is rapidly installed – minimising surveillance downtime due to installation – and simple to operate – minimising the learning curve involved.

Intelligent video vendors are tasked with delivering “sensor-grade” video analytics that are appropriate for the mid range market. A suitable intelligent video device would be a DSP-based video sensor capable of stand-alone operation which simply needs to be hooked up and configured, operating independently of other components. This type of architecture allows system integrators and installers to easily upgrade existing surveillance systems with a simple video-in, video-out and remote, over the network, setup approach.

When upgrading an existing basic camera to DVR architecture, video analytics is easily introduced without making any changes to the current setup or the system’s operation by security guards. Adding video analytics to a CCTV system enables the automatic detection of security breaches as defined in the system through its configuration of rules and detection zones, with a graphical overlay of the detection displayed on the matrix/DVR.

Using the sensor’s dry contacts to trigger recording and event tagging within the DVR allows extending the DVR’s recording capabilities by days or even weeks, since only pre and post alarm footage needs to be saved and eventually backed up.

Military test bed

Not Every High-End System is Suitable for Mid-Range. Although there is a clear differentiation between the needs of high-end and mid-range markets as laid out at the beginning of this article, the performance requirements of the “lower” market are no less demanding. Cost effectiveness and system performance, however, can go hand in hand.

Today’s high performance video analytics systems are required to undergo thorough testing. A manufacturer whose products succeed in obtaining military certification and National Labs’ approval demonstrates that they stand the test of time and are cost effective for mid-range markets.

In tandem with the increased interest in video analytics over the past five years, many new companies have emerged in the market. With varying levels of performance and no unified standard, the user would be well-advised to find out how well a system performs and seek installation references.

The military represents the perfect test bed for security products. With diverse locations and challenges such as rain, snow, wind, darkness, animals, cloud shadows, slow-moving or camouflaged intruders and far distance detection, the army is an ideal organisation to test and approve a new technology and provide feedback for improvement.

Many technologies available to the public today originated in military labs and testing grounds. A video analytics system that has passed the harshest tests in the most difficult terrains is obviously suitable for high risk sites. The criteria for success in a military test include a low false alarm rate; high probability of detection; ability to perform well over long periods of time (sometimes measured in years); rapid deployment; and ease of integration.

These factors, along with a bullet-proof design, are all necessary for off the shelf purchasing and installing.

National Labs provides further insight into criteria such as high performance versus cost effectiveness. Its recognition and approval confirms the system’s suitability for a range of applications, from military and high risk facilities to commercial and public locations.

Last, but probably most important, are references from actual system installations around the world. The more sites the system has been installed at, the more diversified is the manufacturer’s experience in dealing with myriad detection scenarios and false alarms reduction.

Sell value, not equipment

While CCTV cameras, monitors, fence sensors, cabling and access control are the building blocks for any surveillance system, video analytics provides value to an existing or new CCTV system.

Thus, the approach to selling video analytics is fundamentally different to that of standard security equipment.

Video analytics is sold to end users and not security advisors because it creates value by saving money from the end users’ perspective. If the VA system costs the end user $50,000 but engenders savings in manpower, loss prevention and damages of $100,000 per year, then this actually represents a $50,000 gain during the first year and $100,000 every year thereafter.

Capable of filtering out irrelevant information, detecting security breaches, alleviating video monitoring tasks and providing security operators with the necessary data to make informed decisions, video analytics is a key component in the value chain of central command centres.

No matter which system architecture they offer their clients, system integrators and distributors should point out how VA enables better surveillance; saves money; and frees resources to be allocated to other urgent security needs.

Propagation and distribution

Along with the proliferation of video analytics systems and their diversified architectures, PC-based embedded software, edge devices, intelligent encoders and IP cameras, distribution channels have also multiplied and become multifaceted.

Distribution of some architectures is very complex and not all are suited to existing distributors/resellers’ sales models. The reseller may often be required to acquire all the components needed by the integrator or end user, e.g. PC, video grab card, storage devices, displays, etc. They may also be required to integrate these components, which could be beyond their scope of expertise and require heavy investment.

Other architectures can be cumbersome with regards to installation, support and maintenance, as they involve dealing with multiple vendors to assemble a particular video analytics solution. For example, a vendor may provide a video encoder or IP camera in which the video analytics is embedded and supplied by a third party vendor.

It is therefore crucial to simplify the distribution of video analytics for existing CCTV distributors, system integrators and VARs. This means choosing the right architecture for your customers’ needs and having the technical know-how to support them through design, installation and maintenance. This expertise is best acquired through frequent training, white papers, case studies and a knowledge base and learning centre created by the vendor to support the distributor’s marketing, sales and support infrastructure.

Many distributors and system integrators are now looking to complement their product offering with video analytics, due to both increasing demand and interest from their clientele and the significant market potential presented by the technology. Customers may understandably inquire about the viability of video analytics as a replacement to legacy IR sensors.

With the great leap forward in performance, reliability and price competitiveness taken by video analytics over the past years, it makes sense to consider replacing numerous sensors with a system capable of not only detecting but also visually verifying and accurately pinpointing security breaches.

Let’s review some of the advantages of video analytics over IR sensors. A key parameter that must be identified is the coverage area provided by the VA or IR sensor. IR sensors are vulnerable to changes in weather conditions, with their coverage range varying according to outdoor ambient temperature.

For example, an IR sensor operating in the winter is able to detect an intruder at a distance of 21 meters, while in the summer this distance shrinks to as little as 10 meters, resulting in over 50 per cent loss in coverage area.

Because video analytics does not rely on comparing heat emitted from different targets, it is not prone to this pitfall. Coverage range remains constant all year long, regardless of changes in weather conditions.

Furthermore, since IR sensors can only define a single detection zone, it is impossible to discriminate irrelevant areas such as public places. Video analytics, on the other hand, allows the configuration of a number of different active detection zones, passive detection zones and inactive detection zones, as follows:

  • The active detection zone (or alarm zone) refers to areas where an alert is required when a certain event occurs.
  • The inactive detection zone refers to areas within the camera’s field of view but where activities are ignored.
  • The passive zone (pre-alarm zone) is defined as a zone where the system is aware of activities but will deliver an alarm only when an activity passes from that zone to an active zone – for instance, it could be a fence separating a public and private area. In this scenario, the public area would be defined as the passive or pre-alarm zone and the private area as the alarm zone. Activities in the public area are ignored unless an intruder attempts to climb the fence to reach the active zone, which would instantly trigger an alarm. This allows harmless activities occurring in proximity of the private area to be ignored, while allowing a higher probability of detection, faster reaction time and fewer false alarms.

System reliability

In addition to detecting intruders, video analytics enables myriad other detection applications including unattended baggage, illegal parking, loitering, object removal, fence climbing, line crossing, wrong direction and more. Some allow concurrent applications with a single camera input.

Tampering detection plays an important part in a system’s reliability. IR sensors allow the most basic tamper detection, ie, if the sensor is being messed with. Video analytics, however, opens a whole new dimension to tamper detection by also providing immediate feedback on the system’s status.

For example, if weather conditions such as heavy fog impede detection, the system alerts on poor visibility. Other statuses include video signal loss, bad video signal, camera position shift and obstructions. Thus, any scenario resulting in malfunctioning is much harder to circumvent with video analytics than with IR sensors.

Until recently, the most frequent objection to video analytics was that it took much longer to install than IR sensors. Thanks to advances in video analytics technology and DSP-based standalone packaging, video analytics now requires the same minimal time and skills to set up, making it more accessible to installers.

With a higher probability of detection, dramatically fewer false alarms and lower maintenance requirements, video analytics is obviously a viable alternative to IR sensors. Since video analytics covers much more area, 35 IR sensors can be easily replaced with four intelligent video encoders or cameras, as was done at a BMW dealership in the UK.

In this project, the installer manages and fine tunes the system with remote IP access, thus considerably reducing his presence on site. No change to the alarm system was required, thanks to built-in support for legacy dry contacts inputs and outputs. In addition, no end-user training was necessary as alarms just pop on screen, bringing immediate attention to any ongoing security breach. In this project example, the accuracy and reliability of alerts were greatly improved, and instant alarm verification was introduced with real time and recorded video.

The above installation scheme has also been deployed at construction sites in the US, where operators warn off intruders before they can cause any damage or steal anything. This trend is leading to a radical change in the security market, transforming it from a reactive method of operation to a proactive approach, allowing full implementation of the four Ds of physical security: deter, detect, defend and defeat.

What impact on installers?

For installers, video analytics provides a better price/performance ratio, allowing them to reduce expenses by spending less time on site and minimising maintenance calls. Video analytics can help you gain a competitive edge by differentiating your business and enhancing customer satisfaction and loyalty.

For end users, video analytics helps to reduce total cost of ownership, enhance security effectiveness, through better detection performance and reliability, and automate surveillance by delivering immediate alerts with video verification in time to prevent damage.

For consultants, video analytics can be applied to a wide range of customers. Its superior level of accuracy and reliability means less design constraints, and allows using existing and new solution architectures. Video analytics is also future-proof: it bridges to a converged IP architecture once the customer is ready.

In summary

“Sensor-grade” video analytics architecture is most suitable for the mass market. It is flexible, cost effective, simple to install, and allows remote setup and maintenance. It is also widely tested and approved both by the military and National Labs, and has a very large installed base.

Video analytics has many advantages over prevailing sensors: it provides better coverage, has more applications, is harder to circumvent, takes the same time to install, requires less maintenance, is more resistant to weather conditions, and is cost effective.

Video analytics is gaining critical mass. Now is the time for distributors, system integrators and installers to decide whether they are going to catch the train or not.

The decision to come on board is simple enough to make, since the figures speak for themselves. The critical questions are which video analytics vendor to choose, which has the best architecture, which can you depend on for reliable products, support, marketing and sales strategy? There are market leaders and there are market followers. Make your choice an informed one.

Keep in mind that video analytics provides value to all your customers. Video analytics reduces the costs associated with extra manpower, loss of goods and damages ? and you make more money by offering your clients exactly that.

Leave a Reply

avatar
  Subscribe  
Notify of

Sign up to free email newsletters