Public security threats along with increasing security demands from users continually pressure security manufacturers to create ever improving technologies. International news constantly peppers the public consciousness with reminders of present dangers. The public wants to know – can these events ever be prevented?

Video surveillance and intelligence technologies

For everything from safe city projects to home protection, developments are underway all over the world, with their relative obstacles. Safeguarding family members, homes, property, and vehicles are the most common applications using these emerging technologies. But the number of video surveillance cameras used in any given application, with their various networks and demanding resolutions, has increased dramatically with an equal increase in challenges. An ability to fast-track through oceans of data to find the small amount that is actually critical is now in high demand.

On the security side of video intelligence technologies, increasing customer safety has been the primary driving factor. On the business side of video intelligence technologies, increasing customer satisfaction has been the primary driving factor. Here, both service and safety are critical. Facial recognition technologies will not only address these issues but will improve on our abilities to overcome the challenges involved. More than just speed up or enhance current methods, facial recognition will open up new avenues for security and new realms of data wherever it’s needed.

Security industry difficulties and requirements

Identifying suspects during post-event investigation poses many complicated difficulties

Identifying suspects during post-event investigation poses many complicated difficulties. Challenges include everything from reviewing footage from scores of cameras, comparing two face images of the same person with a 10-year age gap, and extracting the 5% (or less) of useful information from massive amounts of data, to implementing pre-event alarms to prevent incidents and managing increased video storage challenges due to increased resolutions. And likely a lot more, too.

These obstacles can prove paralyzing for standard video surveillance systems, even with up-to-date equipment. The sheer amount of information being processed bogs down system hardware. This is where more data-heavy solutions are needed, precipitating the development of deep learning and artificial intelligence products. Along with those new technologies, facial recognition is undoubtedly one of the most important right now.

Deep-Learning benefits for intelligent video surveillance

Conventional intelligent video surveillance carries very strict requirements for a scene’s background. For human facial recognition process, two key steps are required. The first one is Feature Extraction, where the features are designed by humans and are always subjective. This method performs well in very specific environments, and subtle changes in lighting, environment, etc., sacrifice accuracy. The second step is “Classification Learning”, which uses shallow learning and analysis. Far from ideal, the results here directly restrict the breadth and depth of intelligent applications.

Deep learning is intrinsically different from other algorithms. The way it solves the insufficiencies of traditional algorithms is encompassed in a few concepts.

Data processing and storage

Deep learning is very similar to the human learning process and has a layer-by-layer feature-abstraction process

The algorithmic model for deep learning has a much deeper structure than that of traditional algorithms. Sometimes, the number of layers can reach over a hundred, enabling it to process large amounts of data in complex classifications. Deep learning is very similar to the human learning process and has a layer-by-layer feature-abstraction process. Each layer will have different “weighting,” and this weighting reflects on what was learned about the images’ “components.” The higher the layer level, the more specific the components. Simulating the human brain, an original signal in deep learning passes through layers of processing; next, it takes a partial understanding (shallow) to an overall abstraction (deep) where we can perceive the object.

Deep learning does not require manual intervention but relies on a computer to extract features by itself. This way it is able to extract as many features from the target as possible, including abstract features that are difficult or impossible to describe. The more features there are, the more accurate the recognition and classification will be. Some of the most direct benefits that deep learning algorithms can bring include achieving comparable or even better-than-human pattern recognition accuracy, strong anti-interference capabilities, and the ability to classify and recognise thousands of features.

Artificial Intelligence in security

With deep learning technology, the average accuracy of facial recognition increases significantly. Hikvision has demonstrated its ability to advance ahead of the competition. Its advantages in Artificial Intelligence in security have come about by having its own excellent algorithm development team and using the most powerful GPUs in our computer platforms. The results include winning in multiple International Deep Learning algorithm competitions, such as ImageNet, PASCAL VOC, the MOT Challenge, and more.

Achieving the best results come from a combination of having the right hardware and developing algorithms that can perform the computational requirements. Hikvision has the industry's first data computing platform with best-in-class hardware, and completely independent algorithm team.

Hikvision’s facial recognition solution

Hikvision’s Face Library can generate blacklists and whitelists to trigger alarms and automate entry/exit authorisations in a range of applications

Hikvision’s facial recognition solution pushes the limits of what is possible with today’s advancements in recognition abilities. This solution detects human faces and recognises attributes create pre- and post-event face contrasting. With immense capabilities for data storage behind this solution, Hikvision’s Face Library can generate blacklists and whitelists to trigger alarms and automate entry/exit authorisations in a range of applications.

Hikvision embeds its algorithms within the cameras and recorders, so no PC server is needed. This means less delay in transmission, decreasing the burden on back-end devices. More than just efficiency, Hikvision holds a high success rate – the face-capture rate can reach 95% accuracy, with face-comparison accuracy rate up to 98%. Multiple models and types are available, with a wide variety of front-end or back-end combinations to meet multitudes of scenarios and environments. And with world wide support, all human faces are supported.

Enhanced security for multiple applications

Systems integrators will see a wide range of applications becoming possible with this facial recognition solution. It can be applied in banks, luxury shops, and hotels, to recognise VIP customers and initiate specialised services. Hikvision’s solution can be applied in security scenarios, like police stations, customs, international airports, government installations, and much more, to improve security and investigation efficiency.

Download PDF version

In case you missed it

What are the obstacles to adoption of mobile credentials for access control?
What are the obstacles to adoption of mobile credentials for access control?

Using a smart phone as an access control credential is an idea whose time has come – or has it? The flexible uses of smart phones are transforming our lives in multiple ways, and the devices are replacing everything from our alarm clocks to our wallets to our televisions. However, the transformation from using a card to using a mobile credential for access control is far from a no-brainer for many organisations, which obstacles to a fast or easy transition. We asked this week’s Expert Panel Roundtable: When will mobile credentials dominate access control, and what are the obstacles to greater adoption?

How to choose the right security entrance for effective customer security
How to choose the right security entrance for effective customer security

Security and systems integrators across the nation are recommending and providing long-term security solutions to their customers. But when it comes to physical security entrances, integrators can easily fall into the trap of simply fulfilling an end user’s exact request without much pushback. Why? We believe the complexity and variety of entrances available makes it difficult to consult on the best solution, but also because there are a lot of assumptions at play. 1) Ask questions to determine the correct security entrance solution There is confusion in the security industry on the meaning of the word, “turnstile.” End users, when requesting a solution, tend to use the word “turnstile” to describe anything from an old fashioned, 3-arm turnstile to a high-tech optical turnstile to a security revolving door. We encourage security integrators to ask questions to discover how their clients want to mitigate the risk of unauthorised entry or “tailgating.” This can help determine the correct security entrance solution to meet the end user’s goal and budget. By asking the right questions and offering true solutions, you can enhance a relationship built on trust and consultation leading to potential repeat business. Below are four physical security goals—crowd, deterrence, detection, and prevention—accompanied by the type of “turnstile” and its capabilities. This breakdown can help the integrator to confidently address an end user’s request for a “turnstile,” and then recommend a solution that truly fulfills their security goals. 2) Explore options for crowd control Typically seen in stadiums, amusement parks, universities, and fitness centres, tripod turnstiles are considered a low security solution for crowd management. Designed for counting employees or slowing down high traffic volume to collect tickets or payments, tripod turnstiles are built to withstand the most abusive of conditions. Here’s what security integrators should know about tripod turnstiles: Low capital cost, but high annual operating cost due to needed 24/7 guard supervision Lack of sensors can lead to defeat – turnstiles can be crawled under or jumped over without alarm/notification to guard staff Little to no metrics capabilities available – no sensors or alarms if defeated High throughput, handling 30 persons per minute in one direction Full height turnstiles are a tall, robust solution for perimeter fence lines, metro stations or parking garages 3) Choose an effective deterrent A physical deterrent to infiltration, full height turnstiles are a tall, robust solution for perimeter fence lines, metro stations or parking garages. While full height turnstiles do physically stop tailgating (an unauthorised person following someone in the next compartment), they have no means to prevent piggybacking. Two people in collusion can gain access through the full height turnstile by badging once and then squeezing into the same compartment. Here are some other things to note about full height turnstiles: Low capital cost, low annual operating cost Guard supervision is up to the user Little to no metrics capabilities available – no sensors or alarms if defeated Moderate throughput, handling 18 persons per minute in one direction 4) Ensure your chosen turnstile can detect tailgating A staple in lobby security to accommodate visitors, optical turnstiles utilise complex sensors to detect a tailgating attempt. Most models available today offer sliding or swinging barriers. A very common assumption in the security industry is that optical turnstiles prevent unauthorised entry, which isn’t true. In fact, once the barriers are open, a second user can slip through. Or, in the case of a wide lane for disabled use, two people can walk through side by side. In either case, an alarm is generated and supervision is therefore essential in order to respond swiftly. The cost of 24/7 supervision must be factored into the security budget. Here are some other points to make note of: Moderate capital cost, but high annual operating cost due to need for 24/7 guard supervision Sensors detect tailgating and sound an alarm for post-tailgating reaction, but turnstiles can still be defeated Moderate metrics capabilities available (for example, # times tailgating occurred, passback rejection) High throughput, handling up to 30 persons per minute in one direction 5) Determine prevention tactics for staff and visitor safety The entry solution of choice for Fortune 1000 companies, security revolving doors and mantrap portals completely prevent tailgating due to their working principle, ensuring the safety and security of staff and visitors. Commonly used at employee-only entrances, security doors are an unmanned entrance solution that cannot be defeated; sensors in the ceiling prevent tailgating (following in a trailing compartment). Optional piggybacking detection systems are also available (preventing two people in the same compartment from entering). The benefits of utilising a truly unmanned door are unparalleled: guard staff can be reduced or reallocated, and this entrance offers an ROI of just 1-2 years. Here’s more information security integrators should know about security revolving doors and portals: High capital cost, low annual operating cost due to no required guard supervision Sophisticated metrics capabilities available, allowing the end user to prove the value of their security investment Security revolving doors = 20 persons per minute, simultaneously in two directions; Security portals = 6 persons per minute in one direction Biometric devices and bullet-resistant glass can be incorporated for an even higher level of security As we’ve demonstrated here, “turnstile,” in the eyes of an end user, is a complex term that can range from a low security, crowd control solution to a high security, tailgating prevention entrance. Security integrators need to first accurately determine the security goals of their customers and then break down the “turnstile” barrier of confusion to recommend the best solution for fulfilling those goals.

Why customers should buy products and services from SMBs
Why customers should buy products and services from SMBs

In 1973, a brilliant economist named E.F. Schumacher wrote a seminal book titled ‘Small Is Beautiful:’ taking an opposing stance to the emergence of globalisation and “bigger is better” industrialism. He described the advantages of smaller companies and smaller scales of production, highlighting the benefits of building our economies around the needs of communities, not corporations. In almost every industry or market that exists in the world today, you're likely to find a difference in size between companies. Whether it’s a global retail chain versus a small family-owned store, a corporate restaurant chain versus a mom-and-pop diner or a small bed and breakfast versus a large hotel chain — each side of the coin presents unique characteristics and advantages in a number of areas. Disparity in physical security industry Customers are drawn to products and services from large enterprises as the big names typically imply stability This disparity very clearly exists in the physical security industry, and differences in the sizes of product manufacturers and service providers could have important implications for the quality and type of the products and services offered. All too often, customers are drawn to products and services from large enterprises, as the big names typically imply stability, extensive product offerings and global reach. And that's not to say that these considerations are unwarranted; one could argue that larger companies have more resources for product development and likely possess the combined expertise and experience to provide a wide range of products and services. But the value that a company’s products and services can bring isn’t necessarily directly related to or dependent on its size. In an age where the common wisdom is to scale up to be more efficient and profitable, it’s interesting to pause and think about some of the possible advantages of small- and medium-sized businesses (SMBs). Typically, “small” companies are defined as those with less than 100 employees and “medium” with less than 500. Providing social mobility  Schumacher argued that smaller companies are important engines of economic growth. Indeed, according to the Organisation for Economic Cooperation and Development (OECD), a group of 36 member countries that promotes policies for economic and social well-being, SMBs account for 60 to 70 percent of jobs in most OECD countries. Importantly, SMBs provide resilience in that there are often large economic and social impacts when big companies fail. Smaller companies are better for regional economies in general, as earnings stay more local compared to big businesses, which in turn generates additional economic activity. SMBs are also better at providing social mobility for disadvantaged groups by giving them opportunities and enabling them to realise their potential. Smaller companies are often more innovative, bringing to the market novel technologies and solutions such as Cloud, analytics, AI, and IoT New companies introduce new technologies There's no denying the role of start-ups when it comes to innovation. In the security industry, many new technologies (e.g. Cloud, analytics, AI, IoT) are first brought to the market by newer companies. In general, smaller companies’ products and services often have to be as good or better than others to be competitive in the marketplace. They are therefore often more innovative, bringing to the market novel technologies and solutions. And these companies are also more willing to try out other new B2B solutions, while larger companies tend to be more risk-averse. Customer service Aside from the quality of products and services, arguably one of the most important components of a security company’s success is its ability to interact with and provide customers the support that they deserve. Smaller companies are able to excel and stand out to their customers in a number of ways: Customer service. Customers’ perceptions of a product’s quality are influenced by the quality of support, and smaller manufacturers often possess a strong, motivated customer service team that can be relatively more responsive to customers of all sizes, not just the large ones. A superior level of support generally translates into high marks on customer satisfaction, since customers’ issues with products can be resolved promptly. Flexibility. SMBs have a greater capacity to detect and satisfy small market niches. While large companies generally create products and services for large markets, smaller companies deal more directly with their customers, enabling them to meet their needs and offer customised products and services. And this translates to adaptability, as SMBs become responsive to new market trends. By having a pulse on the market, smaller companies have much more flexibility in their supply chain and can adjust much faster in response to changing demand. Decision-making. Smaller companies are much more agile in decision-making, while larger enterprises often suffer from complex, tedious and lengthy decision-making processes. Communication is easier throughout SMBs, as smaller teams enable new ideas to flow and can solve problems faster. Job satisfaction Employees working for SMBs connect more directly with the company's goals and objectives, which in turn increases motivation and job satisfaction Employees working for SMBs connect more directly with the company's goals and objectives, which in turn increases motivation and job satisfaction. SMBs are also generally more connected to local communities and participation in community activities leads to a greater sense of purpose. Additionally, SMBs have a much smaller impact on the environment, which is increasingly becoming an important consideration for today’s employees and customers. Though Schumacher's book takes a much deeper dive into the large global effects of scale on people and profitability, the general impact of a company’s size on its products and services is clear. It’s important for all players in the security industry to remember that the commitment and dedication to product quality can be found in businesses of all sizes. Ensuring safety of people, property and assets Large manufacturers may catch your eye, but small business shouldn’t be forgotten, as they can offer end users a robust set of attributes and benefits. While all security companies are aiming to achieve a common goal of providing safety for people, property and assets, smaller businesses can provide extensive value when it comes to driving the economy, innovating in the industry, providing quality employment and offering superior customer service.