This year, it is predicted that almost a million minutes of video content will circulate the internet every single second. That is a lot of data – and a lot of potential analysis.

The dramatic growth in the use of this medium is driving the demand for more efficient ways of analysing video data, particularly for CCTVs in the security industry.

“There is a growing need for video analytics,” explains Dr Victor Sanchez, from the University of Warwick’s Department of Computer Science.

Current analytics often sees a trade-off between quality and reliability on the one hand and on the other hand, computational power and storage. So, at the moment, analytics are not widely used in train stations, airports, highways, office blocks, housing developments, retail applications and distribution centres.

“We wanted to investigate a solution that would help anyone looking for efficient real-time analytics. We wanted to design a tool which would provide a ‘second set of eyes’ –  a back-up for anyone with the difficult task of constantly scanning multiple, complex screens streaming security footage from a number of different positions.”

Machine learning

Currently, the application of deep learning analytics – where computers can predict or detect an event, based on the rules ‘learned’ from earlier data – are providing a number of solutions. These methods are either based on Convolutional Neural Networks (CNNs) or hand-crafted feature descriptors.

“CNNs have demonstrated outstanding performance for video analysis tasks,” continues Dr Sanchez. “But these algorithms take a relatively long time to get up and running, because there are many layers to the data. It can take days or even months to train a descriptor for something like action recognition.

“Those methods in existence which do require shorter training and processing times, for example tools which use feature descriptors, are highly dimensional in terms of the extracted features and therefore require vast storage capacities and computational resource. This makes them expensive and demanding systems to run.”

New video analytics

The research team at the University of Warwick set to work investigating a new method of feature descriptor for video analytics.

“We looked at a method which encodes the motion information of a Spatio Temporal support region into a low-dimensional Binary string (STB),” explains Dr Sanchez. “The encoded information for the process is obtained from two motion sources – the optical flow and the temporal gradients.

“Using two motion sources provides rich motion information by considering the pixel intensity changes to create a new data space that disregards the background. The tools can therefore quickly detect differences in pixel intensity that are likely to be an abnormal change in a given picture and alert to a possible security breach.”

Promising results

“We are delighted with the results our model is producing so far,” says Dr Sanchez. “It can be used across a wide range of uses including classification, activity recognition, object recognition, video surveillance, monitoring and video anomaly detection, and it is suitable for use in real-time situations. The biggest advance, however, has been successfully reducing the demand on training time to hours, computational power, storage and memory. It means the tool could now run on something like a mobile phone or a CCTV camera itself – no longer needing a powerful computer.”

The team’s findings show huge promise for the security industry.

Dr Sanchez continues: “There are a number of big advantages for a company wanting to employ this kind of solution. 24-hour surveillance is a demanding job that will still need a human eye, but using an automatic real-time monitoring system in the background would give added back-up.

“In addition, a single operator may have the responsibility of watching many screens, streaming many views simultaneously, but the human eye cannot pay attention to every single camera and so a real-time monitoring solution would help to do the job more efficiently. Plus, the low computation power requirement of the tool would allow it to be run on a mobile phone, making it portable and freeing up security personnel to move around a given area to make physical checks.”

Dr Victor Sanchez

Researcher from the Computer Science department at the University of Warwick.

This invention is now the subject of a patent application PCT GB2018/058103, and the researchers are working with Warwick Ventures to explore licensing opportunities for this new solution.

For more information visit: https://warwick.ac.uk/siplab/research/

The post Advances in video analytics for CCTV appeared first on City Security Magazine.

Using the principles of big data analytics and security intelligence can create a number of benefits and value for security departments and their organisations. This article outlines how the wealth of data captured by security departments through their various activities or systems is being used in novel ways to identify and resolve risks.

Identify risks

The wealth of data which security departments capture through their various activities or systems is being used in novel ways to identify and resolve risks (exceptions) which otherwise would go unnoticed until they manifest into issues. The principles of these novel approaches are sometimes classified into ‘Security Intelligence’ and ‘Behavioural Analytics’.

To share a few examples:

A global bank:  employing  more than 150,000 people with over 100 sites across the world was keen to identify instances where their people were ‘remote’ logging into their IT systems despite being (physically) inside their premises. Such exceptions relate to possible duplication of an identity record, which is a serious threat. As the bank discovered, the ‘best’ (cheap, quick and replicable) way for them to approach this was to apply simple principles of data integration and visualisation across their logical and physical access control systems. By doing so, the security team was notified of any such exceptions in real time. This allowed for an instant investigation and helped the bank mitigate their risk significantly.

A Fortune 500 organisation had multiple reported cases of expensive equipment stolen from different buildings around the main campus area. They suspected the thefts were occurring after hours but analysis of access records from their physical access control systems alone wasn’t very helpful. They had hundreds of people working late at that site on a regular basis so they were unable to identify a manageable number of suspects. But, by applying analytical intelligence to an integrated set of time and attendance data; and physical access data, they were able to resolve this. They first defined a ‘usual’ behaviour of an individual and groups of individuals, i.e. which areas they access the most and at what times. Then they looked for exceptions, i.e. if certain individuals or groups accessed certain areas at times which fell outside their ‘usual’ behaviour.

By doing this analysis, a single employee stood out and his access pattern also coincided with the thefts. The next time the employee entered a new area after his normal hours, the Security Operations team was notified following which a guard was sent to inspect the building. The thief was caught red-handed. This approach not only helped them resolve a mystery but also provided them with a strategy to prevent similar activities in the future.

A highly secure Research and Development organisation spent enormously each year to perform background checks for every person accessing their campus. Reduction of their security budget led them to change their policy such that they decided to perform ‘risk assessments’ on each individual and they re-ran checks only on those who represented the highest risk. However, this simply led to a cut in the frequency of checks and raised their risk significantly. So it was critical they re-defined the way in which they derived their ‘risk assessments’.

They started by factoring each individual’s level of access, the time they had been with the organisation and the time since they last went through a background check. This information was coupled with their known ‘behaviour’ (which areas they access frequently and at what times) to compute a ‘Risk Score’. Background checks were mandated for individuals with a high Risk Score and those who showed a sudden increase in their overall Risk Score. This helped a great deal in maintaining their high security levels (no issues reported since) whilst reducing their operational cost by approximately 85%.

Benefits of Big Data Analytics in Security

There are several other use cases, which highlight the benefits and values which security departments are creating using the principles of ‘Security Intelligence’ and ‘Big Data Analytics’.

To outline a few:

• Site utilisation metrics – to what degree is a site being used?
• Key performance indicators – how well are the security operational teams doing based on their service level agreements?
• Identifying risk indicators such as those around tailgating and unused access cards.
• Impact analysis in case of changes such as change of security policies or existing technologies such as access cards and access control systems.
• Supporting the green agenda by reducing the energy usage in areas which are not used heavily based on the data analysed.

How to succeed with Big Data Analytics in Security

However, all great ideas require a successful execution (implementation) for their ‘greatness’ to be recognised. During this study we learnt the following tenets, which were key to successfully achieving the above:

• Identify the use cases, which should be addressed through the endeavours of ‘Security Intelligence’ or ‘Big Data’. Base these on the experiences of known risks, threats and exceptions.
• Look for extensible solutions that can contribute to the bigger picture if that should become necessary. Scalability and extensibility are easily achieved when out-of-the-box solutions are deployed as opposed to customised ones. This helps organisations protect their investment as such solutions can be geared to handle changes of other third party systems or business processes.
• Partner with systems vendors that specialise in vertical security and connect to applicable systems (such as Access control, logical human resource systems, security devices) in a non-customised/non-bespoke manner.
• Avoid generic ‘Big Data’ solutions from vendors that don’t understand security. Domain knowledge is very important given that one size doesn’t fit all. Domain knowledge coupled with reference-able experience of a solution provider implied cheaper, shorter and scalable implementation.

With the above it’s evident that security departments globally are recognising the opportunity to be a business enabler and are aligning their objectives so their organisations can run efficiently. This is a welcome deviation from the traditional view of security being a reactive and investigative team only which was unfairly labelled as a ‘cost centre’.

Dr. Vibhor Gupta, Ph.D., Technology Lead, ASIS UK (at time of writing)

www.asis.org.uk

Read more articles on Security Technology

Read article: Realtime Advances in video analytics 

The post The benefits of big data analytics in security appeared first on City Security Magazine.

In city surveillance, the primary objective is to identify people doing bad things. Thankfully there are many technologies, such as video analytics, to help achieve that goal. These technologies provide evidence when an event has occurred.

However, consider the situation where the perpetrator is not yet a perpetrator, but rather a suspicious target. They haven’t jumped a fence, entered the property or broken a window, but it is often a precursor to a terrorist event or other types of security breaches.  This type of activity is often referred to as “loitering” and it is a behaviour that can be detected using video analytics.

Loitering is the act of remaining in a particular public place for a protracted time without an apparent purpose. Laws that try to prevent loitering have been put in place, but in most cases these laws don’t stick, as the courts generally rule that they are unacceptably vague and do not give citizens clear guidelines as to what is unacceptable conduct. Although not necessarily a violation, loitering is of considerable interest to security personnel. Loitering often indicates imminent intrusion. It may also be an act of information gathering for a planned future event. Gang activity, especially violent crime and drug trafficking, often manifests itself in the form of loitering. Loitering may also provide advanced warning to an event happening elsewhere, as it is often used as a distraction tactic.

When detecting loitering events, security cameras equipped with video analytics tend to be the most appropriate sensor choice.  This has an additional benefit, with most cities already having existing cameras performing city surveillance roles. In these cases, cameras can be augmented with this capability through the addition of a video analytics server, avoiding expensive sensor replacement and installation. Today’s video analytics also include the ability to track unique targets, and, in doing so, they can determine that a specific person, or vehicle, has remained in an area for a period of time that seems suspicious. This also means these algorithms are not misled by multiple targets that are just passing through the scene.

Different cities, as well as distinct public areas within a specific city, may have uniquely different definitions of what entails “loitering”. As such, loitering algorithms typically provide a set of parameters that can be adjusted based on the scene or the specific needs of a particular customer.  These parameters include items such as:

Minimum Targets

For many applications a single loitering person can be grounds for an alert.  However, in some settings, such as a gang hot spot, loitering becomes more of an interest when it exceeds a maximum number of people in a specific region.

Minimum Loiter Time

depending on the area, or even the view of the camera, the amount of time that a target should remain in an area before it is considered “loitering” can vary a great deal.  A public park area could entail upwards to an hour, while loitering for just a few minutes around critical regions of a bridge may be grounds for alarm.

Regions of Interest

Loitering algorithms typically have the ability to be applied to the entire scene, or just a portion of the scene.  Different regions of interest may also have different loitering parameters.

Target Type

Loitering can be specific to the type of target.  In most cases, loitering is thought to be associated with people, but it can also apply to vehicles, watercraft and even wildlife. Many loitering algorithms have the ability to monitor based on the class of target that is of interest.

Real Speed / Resting times

When applying loitering to a large scene with many targets, the act of loitering may better manifest itself as those objects that are, on average, moving at a slower speed, or coming to rest at a greater frequency.

It should be noted that loitering is not always a forbearer of bad news, as it can also be used to identify positive situations.  Loitering detection is frequently used for marketing purposes, determining the effectiveness of a retail display or digital signage.  Car lots may use these types of algorithms to see which vehicles are most popular. Museums or zoos may use it to measure the interest in specific exhibits.

The loitering algorithm is also the basis for other more specific monitoring situations, including the monitoring of queue length, whereby an alarm is issued when the queue or line length exceeds a length defined by the operator, and crowd detection, which may ignore the duration a target has been in the area and instead be more concerned with the number of objects present, alerting when a threshold has been exceeded or when a rapid size change occurs.

Although loitering does not typically entail an actual intrusion or violation of a policy, it is a valuable video detection algorithm that can give the security provider, or marketing person, insight into potential events that may occur or regions that are drawing a higher level of interest.

Eric Olson

Vice President of Marketing, PureTech Systems

www.puretechsystems.com

The post Loitering detection: identifying suspicious targets appeared first on City Security Magazine.

The current economic climate means that times remain tough for law enforcement agencies. Not only are they forced to cope with ever-shrinking budgets and thinly-stretched resources as manpower levels are reduced, they are also under growing pressure to deliver more with less. In looking to meet these challenges, the variety of data agencies need to hold, from crime records to number plate recognition information to suspicious activity reports, could be their biggest asset.

If properly exploited, this data could provide invaluable insight for agencies in their battle to fight crime while driving efficiencies.

Unfortunately at the moment, agencies are not exploiting the full value of the data they have. The issue is that most law enforcement systems today still work in isolated silos. Data is often held across a vast array of standalone systems. Agencies are therefore often not able to access this data easily, or more importantly, use it effectively.

Changing to big data analytics

The good news is that most agencies understand this needs to change. They realise they must find a way of using the big data they hold to streamline their own operations internally and get a holistic view of criminal activity. A data platform that enables information to be viewed and analysed holistically, whether physically located in an HR system, crime or number plate recognition system, has to be the foundation. Once this is in place, then Big Data Analytics can be applied to start to extract insight and provide value.

Of course, advanced analytical techniques have been used in finance and retail environments for years to achieve efficiencies and increase profit margins. The same techniques can and should be applied to law enforcement to achieve similar results. For example, agencies could be analysing all available data to understand crime patterns and keep officers focused on the top crime prevention priorities, understand where duplication is occurring and target resources most efficiently.

Equally, rather than having to hold daily briefings, agencies could use big data analytics to push briefing information in real-time to officers on the beat depending on their physical location – so they are only receiving information that is actually relevant to them.

Today’s data

However, traditional analytical techniques are often not sufficient in the new online world. Gone are the days when most data was held in nicely structured formats, within relational databases. Today, a lot of it is unstructured text in the form of word documents, transcripts, witness statements or internet chatter. This kind of data is potentially really valuable to law enforcement agencies. Unfortunately, in the past, they’ve had to invest vast amounts of time and resource in manually going through this data to make sense of its content and extract the useful nuggets of information from it.

Today, this is rapidly changing thanks to the latest developments in text analytics. Sophisticated linguistic rules and statistical methods can evaluate text just like a human mind – minus the inconsistency and ambiguity. The latest text analytics technology automatically extracts keywords and topics, categorises content, manages semantic terms, unearths sentiment and puts everything in context.

By applying text analytics, agencies can begin to extract intelligence from unstructured data and turn it into a more structured format which can then be analysed together with their structured data. This is really ‘changing the game’ for agencies as it means they can now exploit all of their data, not just the structured content.

Finding the Answer with Big Data Analytics

From the perspective of the agencies themselves, however, the real ‘value add’ of Big Data Analytics is that they don’t need to know what they are looking for before they start. They don’t need to have to wade through the haystack looking for the needle. With the latest advanced analytics technology, officers don’t have to rely on asking specific questions or running specific queries.

Instead, the analytical techniques will model the data and push information of interest back to the officer or analyst, drawing their attention to relevant content, effectively pushing the needle from the haystack. This can then be processed through standard analysis and investigation processes to determine if it is viable intelligence, effectively converting Big Data into actionable intelligence.

To do this, agencies must start deploying the right technologies to extract as much value as they can from that data. Without the right tools, pinpointing relevant data in Big Data that might potentially be of use would be resource intensive and unaffordable. With the right solutions, agencies can sift out irrelevant information and highlight areas of interest, whether that be to achieve efficiencies or drive preventative policing strategies.

Joanne Taylor

Director, Public Security, SAS

The post Finding the answer with big data analytics appeared first on City Security Magazine.

Today the term Big Data is starting to draw a lot of attention from technology professionals, but behind the hype there’s a simple story.

For decades, large companies have been making business decisions based on historical data stored in a variety of databases. Beyond this critical data, however, is a potential goldmine of non-traditional, less structured and previously unrelated data that can be mined for useful information. Decreases in the cost of both storage and computing power have made it feasible to collect this data – which would have been thrown away only a few years ago. As a result, more and more companies are looking to link data that was previously thought to be unrelated, and to begin applying smart business intelligence analysis to provide quality management information.

Traditional security processes and systems have the potential to generate a huge amount of Big Data, but Big Data in isolation is meaningless. The end goal has to be the provision of risk-based information allowing improved security decision making – based on prioritised, actionable insight derived from Big Data. Based upon information from traditional IT data mining, the amount of data available to be analysed by corporate security departments is likely to double every year through to 2016 – and coupled with other relevant Facility and Risk Management data the task may seem impossible.

Over the last several years a category of cloud- based hosted solution software that provides an integration platform and applications has been created. These systems are designed to integrate multiple security applications and devices and control them through one comprehensive user interface. The software collects and correlates events from existing disparate security, facility management, risk and information systems to enable personnel to identify and proactively resolve situations. The system acts as a central repository for data and manipulates the data to produce actionable information using a sophisticated work rules engine. The integration of applications across security, risk and facility management provides numerous organisational benefits, including increased control, improved situational awareness, and timely and accurate management reporting. Ultimately, these cloud-based solutions allow organisations to reduce costs through improved efficiency and to improve security through increased intelligence.

Typically the central repository software comprises a suite of tools which has six key capabilities:

1. Collection: The software collects data from any number of existing disparate systems.
2. Analysis: The system analyses and correlates the data, events and activities, to identify the real situations and their priority based upon defined work rules.
3. Verification: The software presents the relevant situation information in a quick and easily-digestible format for an operator to review and validate.
4. Resolution: The system provides mitigation actions and step-by-step instructions based on best practices and an organisation’s policies, and tools such as a risk audit tool to resolve the situation.
5. Reporting: The software tracks all the information and steps for compliance reporting, training and, potentially, in-depth investigative analysis.
6. Audit trail: The software also monitors how each user interacts with the system, tracks any manual changes to associated systems and manages reaction times for each event.

A key differential between this Big Data approach and other forms of physical security system integration is the ability for a single software platform to connect security, facility and risk management systems at a data level rather than simply interfacing a limited number of products.

The goal of a Big Data analytics software tool is to provide a risk-based security intelligence platform that allows a stronger decision making process and not simply to gather more data! To do this effectively, the solution must be able to reflect a whole series of company- specific work rules and distil down vast amounts of data into meaningful security intelligence.

So, in the context of security, how can this approach really help?

A true Big Data security platform is based around adding business value, and its ability to link into other business systems allows it to increase overall business performance. The potential impact to a business of a security breach could now be too far-reaching to keep related data previously perceived as unimportant down at the operational level. In-depth and specialised reporting used in conjunction with a robust work rules engine using simple built-in tools can easily generate a variety of instant notifications to alert any number of colleagues to an event, and allow the appropriate resources to be deployed to ensure speedy resolution.

A significant added value of a cloud-based platform of this type is its ability to guide an operator through the process of managing incidents. This is typically put in place to ensure security and operational staff comply with processes put in place to meet company risk management policies, and to ensure compliance with legal frameworks and the requirements of regulatory bodies. Often the need to enforce regulatory compliance is the key value driver of this type of solution. This could be as simple as managing the audit process of access control rights to secure areas or reporting on potential Health and Safety incidents. Alternatively, it could be as far- reaching as managing compliance to Emergency Action Plans, where data taken from an existing access control system is referenced to the requirements of an Emergency Action Plan in order to highlight in real time shortfalls in qualified first aiders or fire marshals.

In Conclusion

Ultimately, a Big Data approach must provide an ability to link vast amounts of security, risk and property-related data and present it in such a way that the security and risk professionals can understand and address risk in real time in a way that reflects their business. It’s a challenge that many organisations will need to take on, particularly in the financial sector and for those other organisations with a high level of security risk and regulatory pressure. Success will come from careful definition of objectives and selection of the right software platform to bring together all of the relevant security, risk and facility data.

David Ella

G4S Technology

www.g4stechnology.com

The post Big data in a security context: a simple story appeared first on City Security Magazine.