The deep learning phenomenon continues to excite the IT world, with computing power now at the level where it can be properly used in practical applications.

Hikvision has been at the forefront of applying the technology in the surveillance industry and beyond. It has already released its first set of products that harness the power of artificial intelligence (AI).

The concept of deep learning takes inspiration from the way the human brain works. Our brain is a very complex deep learning model. Brain neural networks are comprised of billions of interconnected neurons; deep learning simulates this structure. These multi-layer networks can collect information and perform corresponding actions according to analysis of that information.The ability of deep learning to enhance the recognition of human beings makes it a great addition to the security arsenal

Reducing false alarms

In the past two years, the technology has excelled in speech recognition, computer vision, voice translation, and much more. It has even surpassed human capabilities in the areas of facial verification and image classification; hence, it has been highly regarded in the field of video surveillance for the security industry.

Its ability to enhance the recognition of human beings - distinguishing them from animals, for example - makes the technology a great addition to the security arsenal. This is especially relevant in a world where false alarms account for 94%-99% of all alarms, according to police and fire service statistics.

Deep learning for data processing

Deep learning is intrinsically different from other algorithms. The way it solves the insufficiencies of traditional algorithms is encompassed in the following aspects.

The algorithmic model for deep learning has a much deeper structure than the 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 learnt about the images’ 'components'.The more features there are, the more accurate the recognition and classification will be

Feature classification and recognition

The higher the layer level, the more specific the components. Just like 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 it 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.

Conventional surveillance challenges

Conventional surveillance systems mostly detect moving targets without further analysis. Even smart IP cameras can only map individual points on a shape one by one, making it difficult to calibrate some features (e.g. forehead or cheek), thus decreasing accuracy. 

For perimeter security, for example, other technologies can be (and are) used to provide more comprehensive security. But they all have their downsides. Infrared emission detectors can be ‘jumped over’ but are also prone to false alarms caused by animals. Electronic fences can be a safety hazard and are limited in certain areas. Some of these solutions can also be expensive and complicated to install.Frequent false alarms are always an issue for end-users, who need to spend time to investigate each one

Identifying false alarms

Objects such as animals, leaves, or even light can cause false alarms, so being able to identify the presence of a human shape really improves the accuracy of perimeter VCA functions.

Frequent false alarms are always an issue for end-users, who need to spend time to investigate each one, potentially delaying any necessary response and generally affecting efficiency. 

Imagine, for example, a scenario where it’s relatively quiet – a location at night where there are few cars and people around. Even here, there could be 50 false alarms in a night. Assuming it takes 2-3 minutes to check out a false alarm, and that just 3 out of the 50 warrants more attention – say 15 minutes each.

A guard either needs to check the system and look back at the alert, or someone needs to be dispatched to the location and look around, checking if anyone has indeed ‘entered without permission’. In most organisations, these would need to be reported/recorded too, adding to the overall time spent on this ‘false alarm’.

Increasing recognition accuracy

Deep learning, however, makes a big difference. With a large amount of good quality data from the cameras and other sources, like the Hikvision Research Institute, and over a hundred data cleaning team members to label the video images, sample data with millions of categories within the industry have been accumulated. With this large amount of quality training data, human, vehicle, and object pattern recognition models become more and more accurate for video surveillance use.

Based on a series of experiments, the recognition accuracy of solutions using the deep learning algorithm increased accuracy by 38% - applying this to the previous example, that’s a saving of nearly one hour each night. This makes deep learning technology a great advantage in a perimeter security solution, with much more accurate line crossing, intrusion, entrance and exit detection.The value of deep learning technology stretches further than traditional security

Accurate motion detection

The value of deep learning technology stretches further than traditional security. For example, tracking movement patterns of individuals can see if they are ‘loitering’ and a potential threat in the future.

A threshold could be set to five metres radius of movement, or ten seconds of staying in the same place. If the person passes either threshold, an alarm could be triggered. The solution tracks the individual and compares this behaviour to a database to see if it recognises a pattern.

Another application would be in a scenario where ‘falling down’ could be a threat, like an elderly care home. If a height threshold was set at 0.5m and duration time 10 seconds, for example, the solution would be able to see a person falling down (as they go below 0.5m) and might be in trouble (if they ‘stay down’ for longer than 10 seconds).

AI-powered deep learning

The solution uses the parameters set to compare with its database and raise an alarm. With features and benefits like these, it’s easy to see how many smart applications could be catered for by deep learning technology.Artificial intelligence has massive potential

To sum up 10,000-strong R&D Centre is pushing the boundaries of surveillance solutions and bringing even more benefits to them. Artificial intelligence has massive potential, and Hikvision is always exploring new ways to apply this exciting technology throughout the security industry and beyond.

Intelligent search

Hikvision deep learning solutions are available at three levels:

  • DeepinView camera can conduct target tracking, grading and capturing when an alarm is triggered.
  • ‘Traditional’ IP camera using a DeepinMind NVR will add the function of searching intelligently by picture, saving time on searching for targets compared with a regular NVR.
  • DeepinView camera and a DeepinMind NVR delivers a full power solution, with the camera sending the information to the NVR, which can then analyse it. This accelerates recording and false alarm filtering.
Download PDF version

In case you missed it

Unifying the mobile experience: cloud, IoT and the AI evolution of access control in 2019
Unifying the mobile experience: cloud, IoT and the AI evolution of access control in 2019

The industry faces numerous challenges in the coming year. Physical and cyber security threats continue to become more complex, and organisations are struggling to manage both physical and digital credentials as well as a rapidly growing number of connected endpoints in the Internet of Things (IoT). We are witnessing the collision of the enterprise with the IoT, and organisations now must establish trust and validate the identity of people as well as ‘things’ in an environment of increasingly stringent safety and data privacy regulations. Meanwhile, demand grows for smarter and more data-driven workplaces, a risk-based approach to threat protection, improved productivity and seamless, more convenient access to the enterprise and its physical and digital assets and services. Using smartphone apps to open doors Cloud technologies give people access through their mobile phones and other devices to many new, high-value experiencesEnterprise customers increasingly want to create trusted environments within which they can deliver valuable new user experiences. A major driver is growing demand for the ‘digital cohesion’ of being able to use smartphone apps to open doors, authenticate to enterprise data resources or access a building’s applications and services. Cloud technologies are a key piece of the solution. They give people access through their mobile phones and other devices to many new, high-value experiences. At the same time, they help fuel smarter, more data-driven workplace environments. With the arrival of today’s identity- and location-aware building systems that recognise people and use deep learning analytics to customise their office environment, the workplace is undergoing dramatic change. Improved fingerprint solutions Cloud-based platforms and application programming interfaces (APIs) will help bridge biometrics and access control in the enterprise, overcoming previous integration hurdles while providing a trusted platform that meets the concerns of accessibility and data protection in a connected environment. At the same time, the next generation of fingerprint solutions will deliver higher matching speed, better image capture quality and improved performance. The next generation of fingerprint solutions will deliver higher matching speed, better image capture quality and improved performance Liveness detection will ensure that captured data is from a living person. Biometrics authentication will also gain traction beyond access control in immigration and border control, law enforcement, military, defence and other public section use cases where higher security is needed. Flexible subscription models Access control solutions based on cloud platforms will also change how solutions are deployed. Siloed security and workplace optimisation solutions will be replaced with mobile apps that can be downloaded anywhere across a global ecosystem of millions of compatible and connected physical access control system endpoints. These connections will also facilitate new, more flexible subscription models for access control services. As an example, users will be able to more easily replenish mobile IDs if their smartphones are lost or must be replaced. Generating valuable insights with machine learning Machine learning analytics will be used to generate valuable insights from today’s access control solutionsEducation, finance, healthcare, enterprise, and other niche markets such as commercial real-estate and enterprises focussed on co-working spaces will benefit from a cloud-connected access control hardware foundation. There will be a faster path from design to deployment since developers will no longer have to create an entire vertically integrated solution. They will simply add an app experience to the existing access control infrastructure. New players will be drawn to the market resulting in a richer, more vibrant development community and accelerated innovation. Data analytics will be a rapidly growing area of interest. Machine learning analytics will be used to generate valuable insights from today’s access control solutions. Devices, access control systems, IoT applications, digital certificates and location services solutions, which are all connected to the cloud, will collectively deliver robust data with which to apply advanced analytics and risk-based intelligence. As organisations incorporate this type of analytics engine into their access control systems, they will improve security and personalise the user experience while driving better business decisions. 

What characteristics do salespeople require in the physical security industry?
What characteristics do salespeople require in the physical security industry?

A basic tenet of sales is ABC – always be closing. But it's a principle that most professional salespeople would say oversimplifies the process. Especially in a sophisticated, high-tech market such as physical security, the required sales skills are much more involved and nuanced. We asked this week's Expert Panel Roundtable: What unique characteristics are required of salespeople in the arena of physical security systems?

Can microchip implants replace plastic cards in modern access control?
Can microchip implants replace plastic cards in modern access control?

A futuristic alternative to plastic cards for access control and other applications is being considered by some corporate users in Sweden and the United Kingdom. The idea involves using a microchip device implanted into a user’s hand. About the size of a grain of rice and provided by Swedish company Biohax, the tiny device employs passive near field communication (NFC) to interface with a user’s digital environment. Access control is just one application for the device, which can be deployed in lieu of a smart card in numerous uses. Biohax says more than 4,000 individuals have implanted the device. Using the device for corporate employees Every user is given plenty of information to make an informed decision whether they want to use the deviceCurrently Biohax is having dialogue with curious corporate customers about using the device for their employees. “It’s a dialogue, not Big Brother planning to chip every employee they have,” says Jowan Österlund, CEO at Biohax. Every user is given plenty of information to make an informed decision whether they want to use the device. Data capture form to appear here! “Proof of concept” demonstrations have been conducted at several companies, including Tui, a travel company in Sweden that uses the device for access management, ID management, printing, gym access and self-checkout in the cafeteria. Biohax is also having dialogue with some big companies in the United Kingdom, including legal and financial firms. Österlund aims to have a full working system in place in the next year or so. A Swedish rail company accepts the implanted chip in lieu of a paper train ticket. They accept existing implants but are not offering to implant the chips. Österlund says his company currently has no plans to enter the U.S. market. The device is large enough to locate easily and extract if needed, and small enough to be unobtrusive Access control credential The device is inserted/injected below the skin between the index finger and the thumb. The circuitry has a 10-year lifespan. The device is large enough to locate easily and extract if needed, and small enough to be unobtrusive. The only risk is the possibility of infection, which is true anytime the skin is pierced, and the risk is mitigated by employing health professionals to inject the chip. Use of the device as an access control credential or any other function is offered as a voluntary option; any requirement by an employer to inject the device would be illegal, says Österlund. It’s a convenient choice that is made “based on a well-informed decision by the customer.” Aversion to needles, for example, would make some users squeamish to implant the device. More education of users helps to allay any concerns: Some 10% of employees typically would agree quickly to the system, but a larger group of 50% to 60% are likely to agree over time as they get more comfortable with the idea and understand the convenience, says Österlund. Protection of information The passive device does not actively send out any signals as you walk. It is only powered up by a reader if a user has access rightsIn terms of privacy concerns, information contained on the device is in physical form and is protected. The passive device does not actively send out any signals as you walk. There is no battery. It is only powered up by a reader if a user has access rights. With use of the device being discussed in the United Kingdom, there has been some backlash. For example, Frances O’Grady, general secretary of the Trades Union Congress (TUC), has said: “Microchipping would give bosses even more power and control over their workers.” A big misconception is that the chip is a tracking device, says Österlund. It isn’t. “We love people to get informed,” says Österlund. “If they’re scared or apprehensive, they can just read up. It’s not used to control you – it’s used to give you control.”