Introduction
Whether you’re considering the perimeter of a massive urban center like Atlanta or a Network Access Point (NAP), a public exchange center, you’ll need to think big. How big is Atlanta’s perimeter, bound by Interstate 285? Amsterdam, Bagdad, New Delhi, Paris, Las Vegas, Denver, Nashville, Washington D.C. all fit inside Atlanta’s Beltline.
The “perimeters” of regions, cities, urban areas can be physical, virtual, temporary or even visual.
The Japanese Kanji characters for this article’s focus are themselves a path to securing the perimeter:
周 (circumference) + 囲 (surround) = 周囲 (perimeter)
Trace a physical or virtual protection path around the circumference of your soft target, surround it with sensors and you have taken a first, important step towards perimeter defense.
A virtual perimeter can be “drawn” to search for public, social media language that could reveal potential threats. For example, the use of a Social Media Accumulator application like Geofeedia reports the use of violent keywords, names of soft targets or gang-related language (figure 1).
Corporate security and first responders can ingest, analyze and share this data with tactical video, acoustic signature or even vehicle license plate capture within our perimeter. That word, “ingest” may be a new one to some in the security industry, but well used in markets relying on “big data” or applied data sciences.
Legacy approach to security data science:
Collect quality or restrictive subsets of security data within your perimeter for higher assurance of quality alarm processing and response.
Progressive approach:
Ingest all available security data and use cognitive, specialized processes to efficiently process alarms and response scenarios.
The ability to use our progressive “ingest all data-analyze later” approach is also recognized as critical to national security:
“We can only query against that which we have collected…if someone has never made a ripple in the pond…in a way that would get their identity or their interest reflected in our database…there will be nothing [to] show up because we have no record of them. ‘Data’ seems a dry and boring word but, without it, we cannot understand our world and make it better.” -- James B. Comey, Director, Federal Bureau of Investigation
To see or not to see: Perimeter Spectral Intelligence
The modernization of facilities like this will need to consider object detection and recognition capabilities provided by today’s video analytic platforms. In addition, leveraging audio and space detection technologies by sensors at the same locations as network cameras would contribute to significant facility security and energy costs. Daylight, low light, thermal imaging and thermal gradient imaging and video surveillance devices, when used at strategic locations, will potentially improve facility security status.
One designer of critical infrastructure facilities in the energy, mining, chemical industries stated the needs succinctly: “It is imperative that warehouses, administrative buildings, workshops and storage tankers are under 24-hour surveillance. This ensures that our customer has complete control of the entire site at all times. Any movement after working hours will automatically trigger an alarm that security officers can react to immediately.”
Creating images based on the heat that is generated by any object, person or vehicle and requiring no additional light source, thermal imaging network cameras are ideal for securing perimeters, dark regions, and areas obscured by foliage.
Both thermal imaging and thermal gradient technology cameras not only perform object recognition in the absence of visible light, they can determine the status of machinery, computers, and latent heat signatures. In other words, it is possible to detect if a vehicle recently driven is left abandoned, depending on the outdoor ambient temperature. With thermal gradient display, it is possible to more accurately detect objects, as well as machines and power transformers nearing thermal failure (figure 2).
Perimeter Protection Ecosystem
Imaging devices represent one sensor class of many available. Edge sensors and physical barriers also can work together to filter events, reduce false alarms and even act autonomously. Some of the categories with examples are listed below: (figure 3)
> Physical boundary with Perimeter Intrusion Detection System (PIDS) sensor(s)
– Perimeter sensors only (electronic fence and barrier protection)
– Perimeter sensors with virtual perimeter video surveillance
> Virtual Perimeter with video surveillance
– Single algorithm
– Multiple algorithms with logic and event
– Intrusion detection across multiple fixed
– Object recognition, metadata search, central station alerts with deep learning
– Object recognition and target acquisition
– Loitering
– Platooning (autonomous fleet vehicles)
> Space Protection
– Stand-alone without alarm verification
– Space protection plus verification by alternate sensors
– With alarm verification and logic by embedded application running in-camera analytics platform
– Laser-based with automated control of fixed or PTZ IP Camera via embedded application
– Wide area surveillance
– Library of multiple object classes for recognition
– Multiple detection zones sizes, angles, classes
Video Analytics
Both visible and thermal imaging cameras can perform long range detection and take important steps toward achieving a cost effective perimeter defense system. With embedded applications, objects around the facility perimeter may be automatically detected by a thermal imaging camera and displayed at maximum optical zoom via a controlled PTZ camera. These systems are designed for real-time threat detection and tracking, while simultaneously providing a high degree of situational awareness to first responders and security personnel. They also provide improved performance over traditional video surveillance cameras in the following use cases:
- Long-range detection
- Wide-area surveillance
- Day/night surveillance
A vital part of any Perimeter Detection Strategy is the use of a Security Operations Center (SOC), or a central location from where staff manages or monitors access control systems, video surveillance, and possibly controls lighting, alarms, and vehicle barriers for remote site(s) using telecommunications, security and data processing technology. With the use of systems like Long-range and Wide-area detection and surveillance, the electronic systems within the Security Operations Center at the site detect suspicious activities allowing the staff to assess the activity and initiate the appropriate response. Two conditions where these detection systems reduce false alarms and optimize SOC response are shown below:
Perimeter protection and intrusion detection systems and methods have become essential parts of most security and surveillance systems. Perimeter intrusion detection systems (PIDS) have been deployed in a significantly large number of vertical markets. The core detection technologies that are currently used in PIDS include sensing cable, taut wire, vibration sensors, microphonic sensors, microwave sensors, seismic sensors and many other traditional solutions. The use of object detection and virtual boundary video surveillance system technologies has increased due to lower cost of deployment. The potential of further false alarm reduction by embedded camera analytics working with passive infrared and laser space detection (figure 4).
In this Perimeter Defense Strategy, the camera embedded applications platform running network module allows IP-based laser space detection sensors to send commands directly to the PTZ camera. Freely rotating in three dimensions (X-Y-Z axes), the camera sensor actively changes orientation and zooms based on alarm outputs from the sensor. Other commands are supported, like triggering the camera to record, or trip output contacts that turn on lights, alarms, or send alerts. The commands are sent directly to the camera, independently of the video management system (VMS) for lowest possible latency. Filtered events and streaming video sent to the VMS is more actionable for two reasons:
- Verification of space detection alarm by network camera’s embedded video motion detector
- PTZ camera or digital preset of fixed, Full HD network camera are automatically acquired, streaming the video data of the alarm location to the SOC.
In any facility, entry controls play a significant role in a Perimeter Defense Strategy. The automated space protection system is also applicable to personnel entries equipped with turnstiles. The laser detection system moves the PTZ camera to the appropriate turnstile lane, recording entering and exiting personnel faces. Potentially, facial recognition software at the VMS can automatically verify both an employee’s and visitor’s last location, critical in the case of emergency.
Person, object and vehicle video analysis tools, thanks to modern data science are well on their way gaining widespread market acceptance. Some applications entered the open source communities in 2016. This will allow more users to model and analyze the motion of objects in security videos. By overlaying simple dynamical models directly onto video, audio, and metadata, also known as Digital Multimedia Content, everyone from operations managers in busy airports to entry screening security officers may see how well a model matches the real world.
The “intelligent” perimeter can benefit from cloud-based deep learning and image processing architecture, distributing the image processing task between a camera edge device carrying an “agent” and a server. This architecture eliminates the need to stream video to the cloud, useful in low bandwidth environments and allowing the numbers of cameras to scale and be in geographically diverse locations, connected to the analytics service. This pay-per-use model encourages the growing breed of central monitoring operators taking advantage of analytics for video verification and false alarm reduction (figure 5).
PIDS and the Defense Strategy Cycle Standard
The NERC CIP-014 Guidelines and Technical Basis document is often used as a reference guide for target hardening of critical infrastructure facilities. Recognizing the advancements in video surveillance and video analytics in Perimeter Defense, the strategy cycle has been enhanced. In older perimeter strategies, video surveillance impacted only detection. With the updated strategy, video surveillance, video analytics, and PIDS play a role in the detection, assessment, and communications.
Perimeter Defense Strategy Cycle
Deterrence Measures: These are visible physical security measures installed to persuade individuals to seek other, less-secure targets. Examples include perimeter signage, the principles of Crime Prevention Through Environmental Design (CPTED, see www.cpted.net) and fencing/walls/gates/natural barriers. Some facilities use public view monitors to visually inform visitors about property surveillance, but users are investing in other means of theft deterrence.
Detection Measures: This is the category where monitoring at the SOC through access control systems, video surveillance, intrusion detection, lighting, and vehicle gates and barriers are performed. In addition, modern systems add sound detection through acoustic signature technology (such as by Sound Intelligence and Louroe Electronics) that detect the sounds of movement and or gunshots/explosives and notify the Security Operations Center.
The ingesting and processing of all the security data, applying a rule set and creating predictive response models is the future of not only critical infrastructure facilities but smart cities.
Delay Measures: Physical security measures installed to delay an intruder’s access to a physical asset include vehicle barriers, landscaping, structural, Critical Component Protection (like power transformers), multiple layers of delay and finally Buffer Zone Protection.
Assessment Measures: Are the alarms legitimate? Here’s where video surveillance, analytics and many of the aforementioned false alarm reduction technologies take hold and establish the “veracity” of alarm data.
Communications: The ability to transport alarm/video signals, audio, and data to the SOC and the growing population of mobile users truly defines where we can install sensors for our Perimeter Defense Strategy. This is where some of the biggest IoT trends like Mobile Edge Computing (MEC) and Fog Computing will support an evolving new user: The Mobile Security User.
MEC offers perimeter security application developers cloud-computing capabilities and an IT service environment at the edge of the mobile network.
The perimeter security mobile-optimized environment is characterized by:
- Proximity of mobile devices to the protected perimeter or controlled entry/exit locations
- Ultra-low latency of alarm processing and confirmation, alarm reporting and tactical video
- High bandwidth availability for Tactical Video on Demand (VoD)
- Real-time access to available spectrum, wireless radio status, network information, device health, firmware revision levels and cyber security status
- Location data; location-based services
Whether you are a first responder or critical infrastructure facility manager, corporate security, richer video content means more sophisticated transport of Digital Multimedia Content (DMC) that could save a life. How long will the first responder have to wait for this content to reach their consuming device? This period of time is the “latency” which should be kept at a minimum at all costs. Also to reduce latency, the use of Mobile Edge Computing (MEC), an accelerating trend in the Video on Demand (VoD) industry will reduce that latency and get critical data to a first responder’s smartphone fast.
Perimeter use case: protecting the outdoor perimeter using MEC:
Our MEC use case first analyzes live video streams from cameras viewing both perimeter and space surrounding a facility. The distributed video analytics solution triggers events automatically, including crowd movement, abandoned or missing objects. Tactical video is streamed to security personnel responding to the analyzed feed in real time and with low latency. Perimeter devices send low bandwidth video metadata to the central operations and management server for database searches, mining, and deep learning.
Perimeter Use Case: Securing the interior of a soft target, using MEC:
Many of our security applications depend on GPS for real-time situation awareness of people and objects. MEC offers Mobile Device location in real time and in a passive way without dependence on GPS, providing perimeter security enhancements in stadiums, airports, transportation hubs, theaters and other areas of public assembly.
Response
These are the immediate measures taken to assess, interrupt, and/or apprehend an intruder, including both automated and manual responses to video surveillance and intrusion detection alarms.
The future perimeter
Just because you’re connected, it does not mean your system is integrated. Sharing data is key.
The next leap forward in our Perimeter Defense Strategy does not restrict our video surveillance or PIDS sensors into a predictive cycle, but organizes our assets by function and permits valuable, actionable data to a “smart security ecosystem,” where data is the real value.
With devices, instruments of perimeter defense and sensors working together and security data being both unstructured and structured, we no longer have to find dedicated security budget funding for a PIDS.
It may well be wasteful to install a device that is not leveraged across the enterprise or multiple users for the protection of soft targets or cities. This is the age where your perimeter shield can be virtually anywhere and deliver ubiquitous protection.
About the Author: Steve Surfaro is the Industry Liaison for Axis Communications. He also serves as Chairman of the Security Applied Sciences Council for ASIS International and Vice Chair of the Security Industry Standards Council.
