In the age of Big Data, IoT (Internet of things), video analytics, deep learning and everything in between, the ability to effectively generate valuable information from raw video becomes a necessity. With the right tools, the volume of data analyzed, the accessibility across an organization and applications for security, safety, operations and beyond, the potential of video surveillance technologies can finally be realized.
The use of video surveillance by law enforcement, security agencies, investigators, public safety, transportation and safe cities is constantly growing. Much has been said about the imminent explosion of video data and the challenges it brings. For video data to be considered useful information that is worthwhile reviewing, the video data must be first tagged, organized, searched, combined, filtered and finally processed into a form that enables organizations to act upon it in a productive way.
The video surveillance industry has experienced major changes in recent decades. From analog to IP, exceptional improvements in quality and resolution, cost-efficient and better storage capabilities, video management systems and the ability to choose between cloud or local architecture, have all contributed to a massive increase in the number of cameras deployed.
Still, we are faced with perhaps the most fundamental challenge: There is more video than eyes to watch.
How can organizations justify the investment in video solutions if the data collected is not being used? It is unfortunate since the video data collected contains valuable information in a multitude of areas and can be used to transform video surveillance from a cost-burden tool to a highly-valued asset, in all types of environments and across numerous verticals.
Video data can be analyzed into three main categories: appearance, dynamics, and statistics.
- Appearance focuses on the ability to accurately analyze what an object of interest (e.g., person, vehicle) looks like and then identify that object within the video.
- Dynamics focuses on the ability to understand patterns of movement, trajectory, speed, direction and more.
- Statistics adds a layer of information that enables understanding the video data in a deeper way, particularly how it can influence an organization’s processes and business decisions.
On top of these three categories is the important ability to then present these layers of information in a time-effective manner. This is where a technology such as Video Synopsis plays an important role. Video Synopsis technology was invented by Prof. Shmuel Peleg of The Hebrew University of Jerusalem, Israel, and is being developed under commercial license by BriefCam, Ltd.
Basing video surveillance technologies on the capabilities of Video Synopsis enables reviewing hours of video in just minutes, by displaying objects in a scene simultaneously, even if those objects entered the scene at different times. Finally, time and space constraints are broken. The massive volume of video data can be viewed in only minutes.
Video Synopsis has been recognized by the industry as an extremely useful tool to review huge amounts of video data effectively, delivering the unique ability to turn video into actionable information by analyzing raw video data and separating dynamic objects from static background. The extracted moving objects are tagged according to classifications such as color, size, date, time and duration, and stored in a database. The background is also analyzed, tagged and stored. Common examples of refinements and classifications include presenting only vehicles in a specific color, only people riding a bicycle, trucks going in a certain direction and more.
Video Synopsis literally makes a long story short. It does this without the loss of context or content, applying full object-tracking capabilities for maximum accuracy. When dealing with video review, the importance of making a long story short is a game changer. It is essentially about eliminating information that not relevant to an investigation and getting to the point quickly. It enables anyone required to review video to accomplish much more both in quantity and quality of investigation related performance.
How are these capabilities being applied in the field?
More and more police departments, law enforcement and security agencies are implementing Video Synopsis-based solutions into their routine investigation workflow, where they can achieve faster time-to-target and maximum accuracy of investigation results. The more innovative departments use Video Synopsis to adopt a proactive approach to security and investigation and have even successfully learned to leverage the use of video data to improve overall residential safety and quality of life within their region.
Real examples from this field vary greatly and include finding stash spots of a closely followed drug dealer by applying a heat map function on the relevant video footage. Then, using the same concept of heat maps, investigators easily pinpointed an area that had become a regular spot for improper garbage disposal and had been severely affecting the nearby residents. In a different case, the ability to manipulate video data according to relevant knowledge regarding a specific investigation case enabled targeting a specific car, which later led to the owner of the car who was then arrested as a prime suspect in a homicide case.
Other use cases and best practices of Video Synopsis solutions include identifying left objects at critical municipal infrastructures, casing incidents (such as banks or bodegas), stalking or domestic violence incidents, sex offenders lurking in parks or schools, as well as in traffic safety for detecting unusual vehicles and/or pedestrian (jaywalker) patterns.
Transportation is another vertical that is increasingly using video surveillance technologies based on Video Synopsis. Video data is leveraged to improve operations management, in addition to the security and safety of passengers and employees. Airports specifically are a complex and challenging environment, as they encompass so many functions on a 24/7 basis. Airports today are essentially small cities, required to monitor and manage traffic, security and operational needs, alongside ensuring passenger satisfaction and continuous improvement of all airport functions.
On top of the effective event monitoring and situational awareness achieved by implementing Video Synopsis solutions, the advanced airports have learned to make use of the huge amounts of video data recorded in other beneficial ways. Best practices include counting rental car returns to ensure accurate payment from the rental companies, managing indoor and outdoor passenger traffic flow throughout the airport, controlling the usage of escalators, monitoring restricted areas and exits, and even optimizing the cleaning services in the airport based on the occupancy patterns of relevant areas and sections.
Liability in such environments is key. Airports find themselves dealing daily with liability claims, as well as the need to review relevant video as part of the claim. Not only does such a process take hours and even days, spending time reviewing video, but many times pinpointing the exact instance of the claimed event is impossible and the airport then gets into the cycle (and saga) of a lawsuit. Applying Video Synopsis enables smart, rapid review and search of the video data, which results in either the acknowledgment that the claim is real (as viewed in the video), or many times the airport will learn that the claim is false. With the video footage, easily available to prove false claims, there is no need for lawyers and endless back and forth procedures.
Hospitals and healthcare facilities share many of the challenges the security and transportation environments face. Hospitals too are becoming mini-cities with the added challenge of functioning under strict regulations and in a constant life-critical emergency mode. Various hospitals have already integrated Video Synopsis solutions into their video surveillance systems, improving overall security and safety across the hospital grounds. In one case, a hospital needed to quickly locate an elderly patient that went missing. The ability to review several hours of video footage in a matter of minutes, on top of filtering the video data to display only what occurred around the patient’s room from which he went missing, enabled quickly finding the patient and ensuring his safe return for his continued treatment.
Looking to add value, hospitals have found that their video data can also be used to monitor and ensure compliance with regulations, which is an integral part of any hospital. By applying Video Synopsis capabilities to video review and analysis, hospitals can monitor behavior and ensure personnel wear the required medical uniforms, wash hands when entering specific rooms, use gloves for certain procedures, and more.
The growing need for reliable video surveillance technologies that add value beyond security crosses industries, verticals, and geographies. This need will only increase as the use of metadata becomes more common in security and video surveillance applications.
Processing a video only once, and then storing the metadata in an indexed database, is the ideal way to enable search results that are returned instantly. Furthermore, within this metadata are the building blocks needed for most video-related applications, from security, safety, and operations to regulatory compliance, business insights and marketing efforts.
In the age when the fastest growing businesses are the ones providing data (rather than hardware) organizations need to rethink their strategy for extracting and analyzing video surveillance data. Organizations facilitating an approach that understands the value and necessity of turning video data into useful information will stay ahead in the game of the increasingly growing data-driven industry.
About the author
Amit Gavish is BriefCam’s General Manager of Americas. BriefCam delivers Video Synopsis solutions, empowering organizations to validate their investment in video by gaining value from video data across all levels of organizations, maximizing security, operations, and business insights. Upon joining BriefCam in 2009, Gavish has successfully introduced a disruptive new technology to the market and established strategic partnerships with ecosystem partners. Since then, he has overseen the growth of BriefCam’s sales infrastructure in North America, serving an ever-expanding client base in the federal, public and private security markets.
