What impact are data analytics having on security?
26 Feb 2020
We live in an era of Big Data. Surrounded by a flood of information, more companies are looking for ways to analyse that information (data) and systematically extract intelligence that can help them operate more efficiently and profitably. The data obsession has extended to the physical security industry, too, where large amounts of data have historically been a little-used byproduct of our access control and even video systems. But the picture is changing. We asked this week's Expert Panel Roundtable: What impact are data analytics having on the security market?
Data analytics are bringing big data to security and changing the way we look at security solutions. In video surveillance, analytics have opened a wide host of applications that customers can use to gather valuable business insights from video data. This not only increases the complexity of the customer solution but brings together stakeholders from departments previously remote in the security design decision. In 2020, a customer-centric design process will be crucial to understand a customer’s business beyond the security or IT department. Keep an open mind while exploring the potential of each new technology and tailor your security design solutions into a catalyst for your customer’s success. Always remember that with big data comes big responsibility. While you increase the value of your customer’s security data, it is just as important to keep that data safe. Analytics and cybersecurity will run together and hold equal importance.
The impact of data analytics is huge in the security market today and expected to continue to expand throughout the spectrum of business applications. Data analytics are applied to virtually every aspect of business intelligence and decision-making related to all things security-related. Better understanding not only what is occurring, but when/how/where presented by data analytics helps answer why it matters, which in turn, leads to better decision-making and putting measures in place to mitigate risks. The most significant impact is the ability to make decisions in real-time, especially as it applies to predictive analytics.
Meeting face-to-face with customers and integrator partners on a daily basis, I often get asked, “how can I do more with my security investments?” This isn’t a new concept. As a society, we’re always aiming to “do more with less” and there is now a significant push to deliver more value with the investment in a security service and/or solution. Analytics is a natural progression for companies that deliver video-related services, such as cloud-based video. The addition of analytics, such as heat mapping, vehicle/people detection or people counting, can take simple video footage and apply intelligence to make data more actionable for security managers. This means the investment in a surveillance camera can become a way to gather more insights and apply those insights meaningfully across an organisation. Analytics offer security integrators an additional way to increase potential revenue and help end users maximise their investments in cutting-edge technology.
At the moment, we are yet to feel a dramatic impact from data analytics, but we are definitely wary about the impact it could have. From an access control point of view, there is huge potential to use the data of any built environment to improve the efficiency of services and to better understand and manage the use of space. Looking forward I believe we are likely to see a lot more use of analytics in security, particularly on CCTV and surveillance systems. There is a great potential to use packets of data to identify people and, although there are questions over privacy in many parts of the world, advances in data analytics are making this a credible outcome. The ability to consolidate and understand previously siloed data (from different security/access control systems for example) is undoubtedly offering many new opportunities from a security perspective.
Advances in computer vision and deep learning have allowed us to develop solutions that can detect and recognise people and objects. Couple this with learning about their past behaviors, and these solutions can understand their potential intent. The data that artificial intelligence (AI) can extract from live video feeds allows us to recognise when an emerging incident would benefit from closer review by a surveillance operator or immediate response from a nearby security professional. Instant notifications when objects or people deviate from norms give teams the time they need to validate and stop bad actors before they do harm — without relying on inefficient and inconsistent in-person monitoring. It also lets us react to operational concerns, such as long queues, and delivers a comprehensive view of crowd demographics, dwell times, and more. This blend of enhanced security and business intelligence is transforming the traditional security market.
While at IBM I was responsible for the analytics on the mainframe worldwide marketing strategy. One of the solutions in my portfolio included real-time scoring. It “scored” the transactional data (for example, credit card transactions) as they were created; the value was more accurate insights as the data was current. For the security market use case, the same is true, the faster we can access real-time situational data the more insight we have in how to respond to an emergency. The key is having a complete picture. You may already have the access control, video surveillance and property management data, but to complete the picture, you also need the data on operational staff. The challenge is that much of that data is still locked in checklists and logbooks, making it old and questionable. With real-time access to the physical operating data, you could respond smarter and faster than ever before.
Today, the datasphere is made up of more than 40 zettabytes of data, according to EMC. That number has doubled every other year since 2010 and doesn’t seem to be slowing. We believe this growth is generating endless possibilities. In a world where the demand for “smart” is on the rise—smart surveillance, smart businesses, smart cities, smart homes—intelligent sensors, including smart surveillance cameras, are collaborating to build a more connected world. These systems are driving quality of life, business efficiency and technology advancement across the security industry. And the data analytics at the heart of these technologies are creating return on investment (ROI) beyond security as well, optimising data management at the edge, identifying anomalies in system components prior to failure to prevent costly maintenance visits for integrators, even enabling end customers like retailers to use foot traffic insights to drive sales.
Video analytics are driving the growth of the security market, resulting in an influx of video and related data. Before, storage systems primarily stored video for playback and review. But today, customers are using their surveillance systems, optimised with AI and intelligent analytics, to deliver important insights. Therefore, there is a greater need for higher bandwidth storage solutions that can accommodate both the reading and writing of video and related data. The proliferation of video analytics has also created a need for high-quality video that demands high-bandwidth storage technologies. When it comes down to it, your AI insights are only as good as the quality of video input. This dilemma has created a greater need for forensic-grade video and reliable video transmission. If there are is video quality degradation, a video analytic will yield erroneous or incomplete information, making results less trustworthy and putting security efforts at risk.
Data analytics expand the usability of the functions and features available in the security market. They are integral in improving accuracy, reducing false alarms and providing better systems overall. For example, analytics enable more accurate predictions, in relation to incidents. As technology has developed, machine learning has become more and more accurate, and the potential now is an increase in efficiency and an overall reduction in false alarms.
Today, we see most industry topics from a security point of view, but tomorrow we'll talk in terms of data and the value of that data and what we can achieve. Tools are being developed now to automate user processes and to augment human operator effectiveness with information. The harvesting of data that's currently left untouched will be a massive catalyst of change within our industry in the very near future.
Data analytics are helping the security industry be more precise in what gets measured. Is that particular person in the video? Did that person scan a product at a register? Did a receipt print? Did all that happen in the correct order? Does the point-of-sale (POS) receipt data look the way I’d expect it to look? If the answer to any of these questions is “no,” then we want to know about it. Without data analytics, tracking this sequence of events would have required a person to manually find and observe each of these transactions--that isn’t a sustainable solution. Now that we’re able to efficiently and affordably gather these analytic facts at scale, we’re able to conduct sophisticated measurements on them. As the data gets better, the questions get more powerful and specific. When the answers come in, the course of corrective action becomes clear.
While basic video analytics have been around for many years, today’s businesses are beginning to recognise that the information these analytics can provide is valuable beyond security. Organisations aim to leverage this data to gain insight into everyday business problems in an effort to improve processes and efficiently mitigate concerns. For example, financial institutions desire the ability to use surveillance cameras for more than just capturing footage — they want to achieve tasks such as enhancing workforce optimisation, improving customer service, or understanding customer behavior. The key to successfully utilising video analytics is leveraging intelligent technology that can simplify and automate a large amount of data to determine what is critical and make it actionable. This is impacting the security market by encouraging manufacturers to extend traditional technology beyond its basic capabilities to help users meet both security and business goals.
Information has value, but creating that value from a flood of data requires modern computer analytics, including new techniques such as AI and machine learning. The security industry has an overabundance of data, which is the raw material of a new source of valuable intelligence. As computer analytics become a more commonplace tool in modern businesses, the potential to transform security data into business intelligence is greater than ever.
- Getting to know Dan Grimm, VP and General Manager of Computer Vision at RealNetworks
- Big wins and the importance of showing up: Insights from SourceSecurity.com editor Larry Anderson
- Setting goals, business travels and radioactivity: Success secrets from Tiandy's John van den Elzen
- Getting to know Jeff Burgess, President/CEO at BCDVideo
Exploring new vertical markets for professional security systemsDownload
5 security lessons for navigating COVID-19Download
Cloud video and smart citiesDownload