A technology poised to transform the physical security market is deep learning, which is a neural network approach to machine learning, differentiated by an ability to train using large data sets for greater accuracy. In effect, the system “learns” by looking at lots of data to achieve artificial intelligence (AI).

Phases Of Deep Learning

I heard a lot about AI, including how it can transform the physical security marketplace, when I attended NVIDIA’s GPU Technology Conference (GTC) in San Jose recently.  

Recognizing images, including video images, is a big focus of AI. In the past, you needed programmers to spend months telling a computer how to recognize an image. In deep learning, instead of programming the computer, you just show it many different images and it "learns" to distinguish the differences. This is the "training" phase. After the neural network learns about the data, it can then use "inference" to interpret new data based on what it has learned. In effect, if it has seen enough cats before, it will know when a new image is a cat.

Factors enabling AI

Deep learning and AI are fast-growing areas for a wide range of uses – physical security is just one. It is all made possible by the coming together of three factors. One is the availability of lots of data. This is the “big data” we have been hearing about; in effect, a proliferation of sensors (including video cameras) has produced a large enough mass of data to enable systems to be trained effectively. The second factor is the development of new algorithms to train neural networks faster, and the third is the availability of computer hardware (specifically GPUs, graphics processing units), that is capable of rapidly completing the involved calculations. NVIDIA manufactures those GPUs and sponsors the annual GTC conference, all about how they can be used more effectively.

“Deep learning is about teaching technology to understand the world around us in a way that is similar to how we understand it”

Deep learning and neural network computing is everywhere. It is now widely available in on-premises computers, in systems embedded in edge devices, and even in the cloud. The edge is particularly important in the video surveillance market, enabling systems to function despite any bandwidth or latency issues that would limit the effectiveness of a central server-based system. Edge-based functionality also limits concerns about the privacy of information, and eliminates dependence on the availability of 3G connectivity.

NVIDIA AI City Initiative

Video analytics applications fall under NVIDIA's “AI City” initiative, which they describe as a combination of "safe cities" (video surveillance, law enforcement, forensics) and "smart cities" (traffic management, retail analytics, resource optimization). Depending on the application, AI City technology must function in the cloud, on premises and/or at the edge. NVIDIA’s new Metropolis initiative offers AI at every system level, from the Jetson TX2 "embedded supercomputer" available at the edge, to on-premises servers (using NVIDIA’s Tesla and Quadro) to cloud systems (using NVIDIA’s DGX).

“AI City applications need an edge-to-cloud architecture,” says Jesse Clayton, Senior Manager, Product Management, Intelligent Machines, at NVIDIA. “Some applications, such as body cameras and parking entrance applications, have to have AI at the edge. But for other problems, you need to aggregate multiple sources of information, such as using AI on an on-premises server for hundreds of video cameras.”

The sheer volume of installed cameras in the world makes video an AI problem – more than 1 billion cameras worldwide by 2020 will provide 30 billion frames of video per day. The existing limitations of current video systems to adapt and function well in real-world conditions point to a need for better technology, as do the traditional shortcomings of video analytics systems. Video systems can achieve "super-human" results, identifying and classifying images using artificial intelligence.

NVIDIA’s Quadro GPU system enables Avigilon network video recorders (NVRs) to search simultaneously across hundreds of cameras to find images that are similar in appearance
NVIDIA’s GPU Technology Conference offered a chance for Avigilon to interact with others focused on AI

AI In Video Surveillance

AI is steadily making its way into video surveillance. Multiple security industry partners are using NVIDIA GPUs to boost the effectiveness of their systems. Many companies highlighted their initiatives at ISC West in April and again at NVIDIA’s GPU Technology Conference. Among them are Avigilon’s Appearance Search and BriefCam’s real-time video synopsis system. Hikvision uses the technology for a six-fold improvement detecting pedestrians in the rain, while Dahua is speeding up its license plate recognition system by five times. Other companies using the technology are UNV Uniview (vehicle classification), SeeQuestor (investigations), Xjera Labs (people and attribute detection) and Sensetime (object detection).

NVIDIA’s Quadro GPU system enables Avigilon network video recorders (NVRs) to search simultaneously across hundreds of cameras to find images that are similar in appearance, such as faces that match an example. The GPU’s fast and efficient processing power, available in a small and affordable form factor, provides a system that is scalable and cost-effective but can run complex algorithms to provide rapid results.

Beyond recognizing objects, the
system can also learn about how
objects interact in the environment,
and look for anomalies

“Deep learning is about teaching technology to understand the world around us in a way that is similar to how we understand it,” says Willem Ryan, Senior Director, Global Marketing at Avigilon. “What seem simple to us in terms of how we perceive the world is complex for a machine to do, but a machine learns faster. Deep learning allows you to teach a machine how to make connections that we make every day. Using GPUs, a system can make assumptions and calculations instantaneously.”

Beyond recognizing objects, the system can also learn about how objects interact in the environment, and look for anomalies or non-typical events. For example, if the system sees a car go onto a sidewalk, it could provide an alert.

How Will AI Develop?

NVIDIA’s GTC conference offered a chance for Avigilon to interact with others focused on AI, and to share Avigilon’s knowledge of the unique AI challenges of the video surveillance market. “This is the heart of the development of AI and deep learning,” said Ryan at the GTC conference. “To be involved and part of this is exciting to Avigilon, and we can expose people here to how AI can be used in a way they may not be familiar with. We have talked to people who didn’t realize how video surveillance happens currently, and how AI is changing it. “

“We want to continue to support the idea of GPU processing and how using it can make video surveillance solutions more effective, and change how people interact with video,” he added. “That’s where we see the impact. There have been challenges we have struggled to overcome in the security industry, and these are the breakthroughs that will help us overcome those challenges. So, we want to be at the forefront and involved in those discussions.”

The impact of AI and deep learning on the physical security industry is only beginning. The full realization of that impact over the next few years will be fascinating to watch.

Share with LinkedIn Share with Twitter Share with Facebook Share with Facebook
Download PDF version

Author profile

Larry Anderson Editor, SecurityInformed.com & SourceSecurity.com

An experienced journalist and long-time presence in the US security industry, Larry is SecurityInformed.com's eyes and ears in the fast-changing security marketplace, attending industry and corporate events, interviewing security leaders and contributing original editorial content to the site. He leads SecurityInformed's team of dedicated editorial and content professionals, guiding the "editorial roadmap" to ensure the site provides the most relevant content for security professionals.

In case you missed it

What Are The New Trends And Opportunities For Biometrics?
What Are The New Trends And Opportunities For Biometrics?

A shift toward touchless devices during the coronavirus pandemic has been a boon to the biometrics sector. Another factor in the recent increase in use of biometrics is lower prices, which are a symptom of a maturing market and of new technology capabilities. Increasingly, integration of biometrics with access control and other security systems is expanding use cases and sales numbers. For additional insights, we asked this week’s Expert Panel Roundtable: What are the new trends and opportunities with biometrics (facial, fingerprint, iris and/or voice)?

Rule Triggering ‘Blacklist’ Provision of NDAA Ready to Take Effect
Rule Triggering ‘Blacklist’ Provision of NDAA Ready to Take Effect

Within days, a rule will take effect that bans from U.S. government contracts any companies that “use” video products from Chinese companies Hikvision and Dahua. The Federal Acquisition Regulation (FAR) rule implements the “blacklist” (or “Part B”) provision of the National Defense Authorization Act (NDAA), which is understood in the security industry as prohibiting dealers and integrators that do business with the federal government from selling Chinese-made video products to any of their customers (even for non-government projects). The rule, which is officially still interim, states: “On or after August 13, 2020, [federal] agencies are prohibited from entering into a contract, or extending or renewing a contract, with an entity that uses any equipment, system, or service that uses covered telecommunications equipment or services as a substantial or essential component of any system, or as critical technology as part of any system.” Federal rules  Within days, a rule will take effect that bans U.S. government contracts any companies that “use” video products from Chinese companies Hikvision and DahuaFederal agencies issuing the rule are the Department of Defense (DoD), the General Services Administration (GSA) and the National Aeronautics and Space Administration (NASA). GSA provides centralized procurement for the federal government. Because the COVID-13 crisis delayed issuance of the rule, the usual 60 days will not be allowed for public comment before the rule is implemented. However, public comments are welcome and will be addressed in subsequent rulemaking. “Telecommunications equipment” refers to equipment or services provided by Huawei Technology or ZTE Corp, both Chinese telecommunications giants. The rule also specifies that it applies to “certain video surveillance products or telecommunications equipment and services produced or provided by Hytera Communications Corp., Hangzhou Hikvision Digital Technology Company, or Dahua Technology Company (or any subsidiary or affiliate of those entities).” Hytera is a Chinese manufacturer of radio systems. Hikvision and Dahua are major international manufacturers of video surveillance equipment. Limits and prohibitions  The rule states: “This prohibition applies to the use of … equipment or services, regardless of whether that use is in performance of work under a Federal contract.” In the industry, this clause is taken to mean that integrators that “use” any of the covered equipment are prohibited from selling to the government. “Use” presumably covers an integrator deploying the equipment in their own facilities and/or selling it to other customers. The rule also prohibits “service … related to item maintenance,” which in the case of a security integrator would include providing service contracts on previously installed systems. Security Industry Association (SIA) The Security Industry Association (SIA) comments: “Due to applicability [of the rule] to uses by entities with federal contracts even unrelated to their federal work, this broad interpretation is expected to have widespread impact on the contracting community across many sectors, as covered video surveillance equipment is some of the most commonly used in the commercial sector in the United States.” Security integrators that do business with the federal government have largely anticipated the new rule and already switched their Chinese camera lines for NDAA-compliant competitors. However, as SIA points out, extensive common uses of the Chinese equipment in various commercial sectors raises additional concerns.   Easing compliance burdens The interim rule adopts a “reasonable inquiry” standard when an offeror (government contractor) represents whether it uses covered equipment. “A reasonable As SIA points out, extensive common uses of the Chinese equipment in various commercial sectors raises additional concerns. inquiry is an inquiry designed to uncover any information in the entity’s possession about the identity of the producer or provider of covered telecommunications equipment or services used by the entity. A reasonable inquiry need not include an internal or third-party audit.” SIA notes that this provision may be aimed at easing the compliance burden by suggesting that contractors only need to inquire based on what information they already possess. The 'blacklist' The new rule covers Paragraph (a)(1)(B), which has informally been referred to as the “blacklist” provision of the NDAA, the John S. McCain National Defense Authorization Act for fiscal year 2019. However, the “Chinese ban” provision [Paragraph (a)(1)(A)] already went into effect a year after the law was signed by President Trump (August 13, 2018). “Part A” covers use of Chinese-made products in fulfilling government contracts. A growing threat Seeking to justify the new restrictions, the FAR rule states: “Foreign intelligence actors are employing innovative combinations of traditional spying, economic espionage, and supply chain and cyber operations to gain access to critical infrastructure and steal sensitive information and industrial secrets. The exploitation of “Telecommunications equipment” refers to equipment or services provided by Huawei Technology or ZTE Corp, both Chinese telecommunications giantskey supply chains by foreign adversaries represents a complex and growing threat to strategically important U.S. economic sectors and critical infrastructure.” SIA has urged a delay in implementing the “Part B” provision, stating: “The federal government estimates that it will cost contractors well over $80 billion to fully implement this prohibition on the use of certain Chinese telecommunications and video surveillance equipment, yet endless delays in publishing the rule now mean that federal suppliers have just weeks to understand and comply with the new rule, which raises as many questions as it answers.” SIA continues: “Federal suppliers across a wide range of industries have increasingly concluded that Part B is unworkable without clarification of the scope and meaning of key terms in the provision, which the rule does not do enough to define. For example, Part B bans agencies from contracting with a provider that “uses” any covered equipment or service. This term is not clearly defined in law or regulation, yet contractors must certify compliance beginning Aug. 13, 2020.” The Part B rule, which only applies to prime contractors, enables agency heads to grant a one-time waiver on a case-by-case basis, expiring before Aug. 13, 2022.

Why Biometrics is Key for the New Era of Convenient Workplace Security
Why Biometrics is Key for the New Era of Convenient Workplace Security

The modern working world has evolved dramatically over the last few decades - from how and when we work, to the places we work from. Widespread internet connection advances, alongside the growth of cloud-based shared working platforms, have not only created the possibility for increasingly flexible working arrangements, but also fueled a desire to do so – particularly among millennials. The preference for flexible working has now created a widespread need for more agile workforces, saddling IT departments around the world with the task to maintain ‘business as usual’ without compromising corporate privacy. With flexible working forecasted to stay for the long haul and passwords increasingly under scrutiny, evaluating alternative secure authentication methods to keep companies’ data and networks safe is important to protect these ‘new normal’ ways of working.   The end of the humble password? A recent report by Raconteur found that the most common method of authentication for securing the digital aspects of workplaces is passwords. Unfortunately, however, between phishing, hacking and simple guesswork, passwords are easily compromised – a problem that is only getting worse, with IT professionals reporting an increase in phishing attacks in the last few years. Once compromised, passwords can be used to enter untrusted apps or websites and, worst and most commonly of all, give rise to even greater data breaches. Between phishing, hacking and simple guesswork, passwords are easily compromised Alongside security concerns, 6 in 10 people worry about forgetting their passwords and, according to a recent Balbix study, 99% of people reuse the same password across different work accounts. This, undoubtedly, is a side effect of the increasingly complex character requirements implemented by many enterprises. This stress and effort leads to frustrated employees, but, more worryingly, forgotten passwords can also cost IT departments millions of dollars a year. In our flexible, hyper-connected world, it is clear then that the humble password is no longer effective. Additional or alternative layers of authentication are needed to help enterprises maintain their workplace security in a more convenient and cost-effective way.   Smarter workplace authentication with biometrics Often, hacking incidents involve the use of stolen credentials. One authentication solution that could bring an end to these large-scale hacking attacks is biometrics, as unique biological traits are extremely hard to steal and spoof. In addition to being a more secure method to authenticate users and prevent fraud in companies’ networks, it is also possible to layer biometric modalities to create a highly convenient and secure multi-modal authentication solution for sensitive areas or information. Spoofing two biometric modalities, such as fingerprint and iris, in the same attack is virtually impossible, but that doesn’t mean this level of security needs to impair the UX. After all, you can put your finger on a touch sensor, while at the same time glancing at a sensor. For businesses, biometrics can be used in a wide variety of use cases, from securing laptops and applications to authenticating employees at secured access and entry points. It can also be used to add frictionless layers of additional security to any aspect of current security systems, such as key fobs or USB sticks, or to access personalized settings or employee accounts when using shared devices, such as a printer system. This way, beyond playing a role in securing the modern workplace, biometrics can also give employees greater flexibility and convenience over how, when and where they work. Privacy and biometrics - explained Many employers and employees worry about safeguarding privacy in the workplace. Considering biometric data is highly personal, it is no wonder, then, that many are concerned about collecting this data for the purpose of workplace security and what liabilities this may expose them to. For businesses, biometrics can be used in a wide variety of use cases, from securing laptops and applications to authenticating employees Employers must adhere to the relevant workplace privacy laws, such Europe’s GDPR, and this duty extends to biometrics, of course. But, providing biometrics is implemented in line with best practice, it can actually protect employees’ privacy far more effectively than its predecessor, passwords. When employers use an on-device approach, their employees can rest assured no one will be able to access or steal their biometric data, as all biometric data is stored and processed on the device - whether that is a laptop, smartphone, USB stick or key fob. Removing the need for data to ever enter the cloud, this also removes the technical and legal complexities of managing a biometric database and, if a key fob is lost for example, all parties can rest assured there is no chance of anyone else being able to use it. A win-win. Precisely because biometric data is so difficult to steal and spoof, adding biometric authentication to end-point devices can considerably reduce data breaches to keep both sensitive employee and corporate data safe and secure. Reimagining workplace security As people work more flexibly, systems are shared more frequently, and attacks get smarter, it is clear to see that passwords alone are no longer enough to secure the modern-day workplace. Adding biometric authentication to end-point devices can considerably reduce data breaches Now is the time to reassess the physical and logical access control infrastructure. To keep personal and corporate data safe, it is crucial to add new and additional authentication methods to the security infrastructure. Luckily, the benefits of biometrics are often far simpler to realize than many enterprises imagine. The beauty of biometrics is its combination of both security and convenience. Compared to other forms of authentication, biometrics offers considerably stronger protection and an enhanced UX that can easily be integrated into existing enterprise security infrastructure – without the need for huge biometric databases to manage or fear. So, whether to replace outdated passwords or as part of a multi-modal authentication system, biometrics can play an important role in pushing workplace security into a new era for both physical and logical access control.