Over the years, video analytics has gained an unfavourable reputation for over-promising and under-delivering in terms of performance. One of the biggest complaints regarding video analytics has been its inability to correctly identify objects in situations which appear trivial to the human observer. In many cases, this has resulted in a tendency to generate substantial numbers of false alarms, while not detecting actual events accurately. This, together with a propensity for complex set-up procedures and much need for manual fine-tuning, has prevented video analytics from becoming a mainstream application deployed on large numbers of cameras.

Machine Learning And Video Analytics

Machine Learning is a well-established field of research that has existed for decades, which is already present in many products and applications. Machine Learning is based on collecting large amounts of data specific to a particular problem, training a model using this data and then employing this model to process new data. With regard to video analytics, one of the most critical problems impacting accuracy is object classification. Fundamental to improving performance is the capability to teach the algorithm to distinguish between people, animals, different types of vehicles and sources of noise at an extremely high level of accuracy.

Deep Learning Algorithms

The recent increased interest in Deep Learning is largely due to the availability of graphical processing units

Until recently there have been minimal applications of Machine Learning used in video analytics products, largely due to high complexity and high resource usage, which made such products too costly for mainstream deployment. However, the last couple of years have seen a tremendous surge in research and advances surrounding a branch of Machine Learning called Deep Learning. Deep Learning is a name used to describe a family of algorithms based on the concept of neural networks. Very loosely speaking, these algorithms try to emulate the functionality of the brain’s neurons, enabling them to learn efficiently from example, and subsequently apply this learning to new data.

The recent increased interest in Deep Learning is largely due to the availability of graphical processing units (GPUs). GPUs can efficiently train and run Deep Learning algorithms, and have allowed the scientific community to accelerate their research and application, bringing them to the point where they exceed the performance of most traditional Machine Learning algorithms across several categories.

Solving Object Classification With Deep Learning

This means that Deep Learning can now be used to solve the most crucial problem facing video analytics – object classification – by collecting many thousands of images from hundreds of surveillance cameras, which must first be manually labelled and classified by a human, into a range of categories that include: person, car, bus, truck, bird, vegetation, dog and many more. To achieve the required accuracy rates, such a vast database must be collected and identified from actual surveillance footage.

A crucial component in achieving and maintaining the high performance of Deep Learning-based applications is the ability to continuously update the models as more data is collected
Basic classification and false alarm reduction are the first applications of Deep Learning for video analytics

A Deep Learning algorithm trained on images collected from YouTube, Google Search and elsewhere on the Internet will completely fail in analysing images from surveillance cameras, due to the difference in viewing angles, resolution and image quality. Once enough images are collected, a Deep Learning classifier algorithm can be trained and deployed as part of a video analytics solution, enabling it to practically eliminate most of the existing causes for false alarms.

Due to GPU requirements in order for the algorithms to run efficiently, video analytics solutions using Deep Learning will initially need to run on a server. A few solutions of this nature are already available and are showing a dramatic leap in performance in comparison to traditional video analytics, with a drastic reduction in false alarm rates and a significant increase in detection accuracy. Concurrently, these new solutions do not require manual tweaking by the user and are essentially plug-and-play, making mass deployment a realistic premise.

Surveillance Applications Of Deep Learning

Basic classification and false alarm reduction are the first applications of Deep Learning for video analytics, but they are by no means the only ones. In the not-too-distant future, we will see Deep Learning enabling as yet not possible video analytics applications, such as identifying objects carried by people, such as a gun, handbag, or a knife, or being able to quickly find people and vehicles with similar appearances across multiple cameras and more.

Over the next few years, we will see a transition of video analytics using Deep Learning running on servers to running inside cameras

Over the next few years, we will see a transition of video analytics using Deep Learning running on servers, to running inside cameras, as powerful, low-cost hardware capable of running Deep Learning becomes more available and a basic function of newer surveillance camera models. This will push the acceptance of video analytics even further, eventually making it a fundamental element of every surveillance camera deployed.

Increasing Accuracy With Updates

A crucial component in achieving and maintaining the high performance of Deep Learning-based applications, is the ability to continuously update the models as more data is collected so that the models increase in accuracy. This will give an advantage to cloud-based video analytics services, since they can collect vast amounts of data from cameras connected to the service, train new models in the cloud based on this data and then push these new models to cameras at the edge. This continuous improvement cycle will be instrumental in helping video analytics fulfil the promise of improving peoples’ safety and security, by giving surveillance cameras human-level accuracy and a comprehensive understanding of the environment.

Download PDF version

Author Profile

In case you missed it

BCDVideo Signs OEM Deal With Dell EMC: Positive Impact For Surveillance Storage
BCDVideo Signs OEM Deal With Dell EMC: Positive Impact For Surveillance Storage

In a significant move for the video security market, BCDVideo has announced that it is set to become Dell EMC’s OEM partner in the video surveillance space. For nearly a decade, the Chicago-based company has been known as a key OEM partner of Hewlett Packard Enterprise (HPE), providing storage and networking technology to security integrators on a global scale. This latest partnership will allow BCDVideo to take their offerings to the next level. BCDVideo Vice President Tom Larson spoke to SecurityInformed.com to discuss the reasoning behind the deal, and how the program will benefit partners, integrators, and end-users alike. Expanding BCDVideo's Product Offering For BCDVideo, the HPE OEM program has been widely acknowledged as a success, allowing the company to leverage a globally recognized brand and provide high-quality, reliable solutions across video networking and access control. Nevertheless, explains Larson, HPE server solutions are primarily suited to large-scale enterprise projects, and are therefore unable to accommodate for the growth in small- and medium-sized surveillance applications. The global collaboration with Dell EMC will allow BCDVideo to open up a broader product offering, building on success in the larger enterprise market to offer tailored solutions to SMEs. Our aim is to look at all best of breed technology to serve the video surveillance marketplace, and that means multiple partnerships” Support For Integrators By leveraging Dell EMC’s sophisticated digital storage platforms, BCDVideo will now be able to offer a more cost-effective solution to integrators, without sacrificing the resilience and IT-level service that BCDVideo is known for. With access to Dell EMC’s expansive global sales and technical teams, the company hopes to expand its reach, all-the-while providing partners with around-the-clock technical support and a five-year on-site warranty. Customers should be reassured that BCDVideo will continue to offer HPE platforms, service, and support. “Our aim is to look at all best-of-breed technology to serve the video surveillance marketplace, and that means multiple partnerships,” says Larson.  “The addition of Dell EMC to our portfolio is a major win for BCDVideo, for Dell EMC, and for our integrators.” The global collaboration with Dell EMC will allow BCDVideo to open up a broader product offering Meeting Surveillance Market Demands At the technology level, assures Larson, Dell EMC’s server offering is well suited to handle the increasing video resolution and growing camera count demanded by the surveillance industry. At the larger end of the spectrum, the company’s Isilon Scale-Out NAS solution can handle tens of petabytes of data, making it ideal for large-scale security applications such as city-wide surveillance and airport security. Dell EMC storage solutions are already proving successful at major international airports including Dubai and Abu Dhabi, each with a camera count in the 1000s.Dell EMC and BCDVideo together are ensuring our customers get the right solutions designed for the surveillance market” For Dell EMC, the new partnership means the ability to expand on this success in the enterprise market, leveraging BCDVideo’s surveillance expertise and high-level customer service to offer tailored solutions for lower-volume applications. Since its inception, BCDVideo has differentiated itself in the security space by providing a high level of IT service to integrators making the transition to IP systems. By combining resources, the partners will be able to service VMS and analytics companies, software vendors, and access control providers, as well as traditional business integrators. Ken Mills, General Manager Dell EMC Surveillance, explains: “Surveillance storage is not just about capacity, it is also about performance and reliability. Dell EMC and BCDVideo together are ensuring our customers get the right solutions designed for the surveillance market.” Accomodating For Growth BCDVideo is well placed to accommodate this anticipated growth. Last year, the company opened a new 51,000-square-foot global headquarters in Illinois, home to 90 separate stations within their Innovation Center where each system is customised according to integrator needs. The new facility allows for expanding business with new and existing partners in the security market.

How To Prepare For Active Shooter Incidents | Infographic
How To Prepare For Active Shooter Incidents | Infographic

This Active Shooter infographic summarises information about trends among active shooter incidents, and outlines how an organization can develop a plan before tragedy occurs, including:   Statistics on the numbers and types of recent active shooter incidents. A profile of common traits among active shooters. How to prepare beforehand, and what to do when the police arrive. How organizational planning ensures maximum preparedness. Pre-attack indicators to look for. Be sure to share this information with coworkers and managers. Awareness is key to preventing active shooter incidents, and to minimising their tragic consequences. When sharing this infographic on your website, please include attribution to  SecurityInformed.com More resources for active shooter preparedness: How hospitals can prepare for active shooter attacks Six steps to survive a mass shooting Technologies to manage emergency lockdowns  How robots can check for active shooters  Background checks to minimise insider threats Gunfire detection technologies for hospitals, retail and office buildings 21 ways to prevent workplace violence in your organisation Non-invasive security strategies for public spaces    

How Should Your Security Company Measure Total Cost Of Ownership (TCO)?
How Should Your Security Company Measure Total Cost Of Ownership (TCO)?

How much does a security system cost? We all know that total costs associated with systems are substantially higher than the “price tag.” There are many elements, tangible and intangible, that contribute to the costs of owning and operating a system. Taking a broad view and finding ways to measure these additional costs enables integrators and users to get the most value from a system at the lowest total cost of ownership (TCO). However, measuring TCO can be easier said than done. We asked this week’s Expert Panel Roundtable to share the benefit of their collective expertise on the subject. Specifically, we asked: How should integrators and/or end users measure total cost of ownership (TCO) when quantifying the value of security systems?