IPVM (IP Video Market Info), the independent source for video surveillance information, does their typically solid job analyzing the TSA report on the SightLogix video analytics system at Buffalo Airport, which we blogged about previously.

Ethan over at IPVM fills in a number of details, and points out that some of the information in the public version of the report has been redacted. Fortunately, the reports (in their full content) are posted on TSA’s Secure Webboard. Every airport is required by regulation to establish an Airport Security Coordinator (ASC) position. The ASC has webboard access provided to them by the TSA. It’s also the location where they go to retrieve Security Directives. We’re told that over 400 domestic airports have access to the full unredacted report.

Unfortunately, SightLogix does not have access to the full report nor does anyone else without clearance to use the TSA webboard. We can however provide additional information that is not confidential regarding the testing at Buffalo Airport. SightLogix equipment went through a 9-month, four-season test by the TSA before being funded at Buffalo Airport. And they are now testing another SightLogix product through the same four-season process. This funding and testing is based upon the favorable results achieved in their test environments.

SightLogix was founded to make outdoor video analytic cameras that can accurately detect targets with low nuisance alerts in the outdoors. We are confident the testing at Buffalo Airport would have yielded a probability of detect at 100% or very close to it, because SightLogix conducts our own testing during qualification and will not leave a site without 100% coverage and margin to spare.

As mentioned in the IPVM report, the integrator on the Buffalo project pointed out that “99% of the alarms were caused by animals and dependant on the size and quantity may or may not be a concern for the airport.”  SightLogix cameras use GPS-based analytics and are georegistered to the scene during initial calibration. This allows the camera to determine the precise location and actual size of all objects in its field of view, allowing users to set very accurate video analytic size filters for eliminating small animals and other objects from sending alerts. This is a key to accurately detecting targets  based upon size, and is the reason that SightLogix systems have the lowest outdoor FAR/NAR in the industry.

At this point in our company’s history, now that we have been able to consistently achieve extremely high levels of detection accuracy, we’ve turned our focus towards cost reduction with our new SightSensor thermal camera model, which has twice the processing power of those originally tested by the TSA at Buffalo and at pricing per-foot on par with less reliable fence sensors and visible cameras with analytics. The additional processing is being used to vastly enhance the thermal image so it more resembles a quality black and white visible image, under all lighting (or no lighting) conditions. The overall cost reduction was achieved by reducing the costs of the electronics and housing and streamlining manufacturing, now that we are in higher volume sales.

Thanks to IPVM for providing a valuable service in clarifying the scope of the TSA  report and presenting the facts. You can find their report here.

An evaluation by the U.S. Transportation Security Administration (TSA) has confirmed 100% video analytics accuracy of the SightLogix SightSensor thermal camera system in the challenging environment typical of many airports.  According to the TSA’s final report, the “evaluation team performed over 900 scenarios of which every alarm instance was accurately reported.”

The TSA established the Airport Perimeter Security test program to address security vulnerabilities at aviation facilities. The goal of the TSA test was to validate thermal video analytic security effectiveness at Buffalo airport, a difficult environment due to its widely varying topology and inconsistent illumination, which is typical of most airport perimeters.

The public version of the TSA report is available here.

TSA evaluators conducted numerous test scenarios to determine the effectiveness of the SightLogix system. The tests were designed to simulate a human intruder attempting to defeat the system by breaching the perimeter detection zone without causing an alarm. The scenarios were distributed throughout several regions of the airport’s perimeter covered by the SightSensors. The evaluators performed randomized scenarios for each SightSensor deployed.

According to the TSA’s final report, “each alarm prompted the system to display the alarm video, location information, nearest camera reference numbers, date and time, and an audible alarm signal.” Additionally, the TSA reported that “SightSensor target tracking capabilities were available and 100% functional throughout the evaluation period.”

According to the TSA, airport personnel reported that the process of integrating the SightSensors into the existing video management system was “smooth and without issue.”

Cost savings were also reported by airport officials. SightLogix on-board image processing, which provides accurate detection in the outdoors, also provides extended detection capabilities. This additional range reduced the number of poles, trenching and communications needed for the airport’s deployment, while exceeding the automated detection area originally specified in the design. The result was a more accurate perimeter security system that met the airport’s available budget.

The use of thermal cameras also provided an optimal detection source for the airport because thermal analytic cameras can detect intrusions that might occur even in complete darkness, removing the need for any illumination.

In the final analysis, the TSA concluded that the SightLogix thermal video analytic solution “had a positive effect on the airport’s perimeter security monitoring and detection efforts.”

We’ve added narration to the SightLogix Automated Outdoor Video System overview, shown below.  Each video panel displays the view from a visible, wide angle and thermal perimeter security camera, while the bottom right shows a PTZ camera automatically steered to zoom and follow detected targets. The target’s location is simultaneously displayed onto the topology map on the bottom left.

Click the video for a guided tour of the system in action.

SightSurvey was designed to help integrators quickly and easily model an outdoor area and perimeter security layout around an actual site.  It’s become one of our most popular design tools.

SightSurvey uses Google Maps to create an intelligent video camera layout on any facility. This lets you address typical security design issues — like camera blind spots and terrain conditions — and determine areas of vulnerability or coverage gaps. By using Google Maps, you can gain a close approximation of the final design requirements even before you visit or walk the area.

With SightSurvey, you can set pole height, view blind spots under the camera, and choose visible or thermal intrusion detection cameras based on the application environment. Even large facilities such as airport perimeters, rail yards or seaports can be designed in a few minutes.

SightSurvey also lets you plan the number of cameras and their optimal placement, and determine a camera’s range for different sized object detection, such as a person versus a vehicle. You can also use SightSurvey to compare the number of cameras required for effective video intrusion detection around your facility.

We’ve put together a brief overview of SightSurvey, which you can watch below.  If you like what you see, register for SightSurvey here.

When it comes to protecting outdoor areas, security professionals often have two main concerns:  Not being aware of the risks that are lurking, and not knowing the place and nature of an intrusion should one occur.

Intelligent video cameras address the first concern by leveraging the inherent strengths of automated systems and people. Smart cameras never tire, can cover large distances, and “see” what the eye would miss. People can then be counted on to make response decisions.

It’s this second concern — the “what and where” of an unfolding event — that has been more difficult to address. Typically, a perimeter intrusion detection system will combine several technologies, including fixed cameras for long-range surveillance and PTZ’s to zoom and follow an object for more detail. The problem is that there’s almost no chance the PTZ cameras will be looking in the right place when an intrusion occurs. Trying to manually locate a detected alarm with a PTZ camera — especially over large outdoor areas — can be like finding a needle in a haystack.

The better way is to use systems that capture GPS positioning data, which is then used to steer PTZ cameras to automatically track and zoom in on intruders, making the target large enough to reliably identify. This information can also be used as forensic recording for post-event management.

You can see this in action in the following video.  Each video panel displays one element of a video intrusion detection system. The left panel shows a visible detection camera; the center panel shows a wide-area camera; the right panel shows a thermal camera, and the bottom right panel shows the PTZ. You’ll see how it is automatically steered to zoom and follow the target, which is simultaneously displayed onto the topology map on the bottom left.

Such automatic control is especially valuable for facilities that monitor large areas with long-range detection cameras.

If you liked this post, subscribe to the SightLogix RSS feed or sign up for email updates.

In our continuing series for reducing perimeter security nuisance alarms we’ve addressed the benefit of image processing and the value of electronic image stabilization. Another requirement for accurate perimeter intrusion detection is for an automated camera to “see” security violations against a background of movement and clutter. Practically speaking, this means the camera must be smart enough to ignore movement from “unimportant” objects. Such movement is also known as non-salient motion.

By definition, non-salient motion comes from objects that always return to the same position within a brief period of time, like branches, trees, foliage, or reflections in water. Alternatively, salient or “important” motion comes from objects that traverse the camera’s field of view, like a person or vehicle.

From a security perspective, salient motion is what matters. It’s activity that likely represents a legitimate intruder entering a secure area which requires investigation. On the other hand, non-salient motion is normal behavior in the outdoors, and must be filtered by the camera or the nuisance alarm rate will reach a level so high they will likely be ignored.

For indoor surveillance applications, such distinctions are irrelevant because background objects are not moving. In the dynamic outdoors the differences between these two types of motion is more important. Think of a typical outdoor scene: a large tree will sway in the breeze within the same general location. The same thing happens with leaves, bushes or reflections in water. Obviously, you want such movement to be ignored, or risk wasting your security efforts chasing after phantom issues.

Filtering such movement is best accomplished with sufficient on-board processing and video memory buffers for making the right determination. This is no easy task, especially considering the amount of data that a camera needs to analyze over a large outdoor scene spanning hundreds of meters. Importantly, when we refer to “video memory buffers” we’re not talking about on-board video storage for future retrieval. In this case, we’re talking about very high-speed memory that’s used for real-time scene analysis.

For example, a large tree takes a period of time to move in one direction and then just as long to return. Cameras that lack sufficient memory will only see the tree moving one way and trigger an alert. On the other hand, if the camera can analyze and store scene information over a longer period of time it can conclude that such background movement returns to the same area and is an “unimportant” object to be ignored.

You can see a real-world example of an intelligent video surveillance camera accurately determining salient from non-salient motion in the following video. Notice how the moving brush and foliage on the left is safely ignored, along with the heat waves coming off the vehicle, while the human targets — which are actually smaller than some of the background objects that are ignored — are accurately detected and alarmed. In fact, this video also shows how intelligent video cameras can “see” what the human eye may have missed.

Without sufficient on-board processing and video analysis memory to filter non-salient motion, such an environment would most likely be overwhelmed with nuisance alerts.

Subscribe to the SightLogix blog and keep updated on this series and other articles related to outdoor security.

Most security folks would agree that deploying reliable automated video systems in the outdoors has been challenging, at best. Generally speaking, video analytics have worked well in controlled environments, such as those in the static indoors. Outdoor security applications are a different story.

Challenges notwithstanding, it is possible to deploy intelligent video surveillance cameras in the outdoors that maintain a high probability of target detection while also addressing the core issue of nuisance alarms. The critical first step  is to focus on image processing in advance of the video content analysis. Such processing resources – often encompassing several digital signal processors (DSPs) –make it possible to analyze the full visual detail of every video frame — inside the camera, at the network edge. In this way, intelligent cameras have the horsepower to clean up the problems associated with the outdoors.

This includes electronically stabilizing the image for camera motion, adapting to changing lighting, fog, rain, snow and sandstorms, and filtering variables such as small animals, blowing debris, trees moving in the breeze and reflections from water.

With such complexity in the outdoors it’s not possible for automated video cameras to accurately determine legitimate targets unless they bring a high degree of image processing to the network edge. When video processing and analysis is performed by video encoders separate from the camera — or by servers in the datacenter — they perform their analysis on a small fraction of the available scene information, at times less than one percent due to preparing data for transmission. Analyzing 100% of the raw scene data directly in the camera as it leaves the imager greatly improves the probability that cameras will accurately detect targets and filter the outdoor impediments that would otherwise trigger nuisance alarms.

Without on-board image processing of sufficient power, the only way to prevent excessive nuisance alarms outdoors is to lower the sensitivity of the system, directly impacting camera range and detection accuracy.

There’s been great disappointment among customers in cases where video intrusion detection systems have been deployed that are not designed to address the outdoor challenges. Substantial on-board image processing in advance of the video analysis is a foundational step, upon which a range of capabilities can be built. These include georegistration of targets, dynamic lighting correction, electronic image stabilization, automatic PTZ steering, and other important security functions. Greater image processing also translates into cost savings through the extended range and such processing affords.

We’ll discuss these additional technologies in subsequent articles. Subscribe to the SightLogix blog and keep updated when they’re published.