Video virtualization and AI for near real-time actionable intelligence needed to identify and mitigate defense and security threats

Rajesh Uppal June 5, 2019 AI & IT Comments Off on Video virtualization and AI for near real-time actionable intelligence needed to identify and mitigate defense and security threats 179 Views

Police agencies are using facial and object recognition technology for counterterrorism operations. Video footage played a key role in finding the culprits responsible for the November 2015 Paris attacks, with a CCTV video at Brussels airport used to pin down one suspect. But, the sheer volume of video content produced makes identifying, assembling and delivering actionable intelligence — from multiple sources and across thousands of hours of footage — a habitually long, laborious process. DoD collects loads of data from satellites, drones and Internet-of-things devices. But it needs help making sense of the intelligence and analyzing it quickly enough so it can be used in combat operations.

Now defense and intelligence agencies are leveraging artificial intelligence (AI) and machine learning to automatically identify video objects of interest. They need powerful artificial intelligence software tools that the tech industry is advancing at a past pace. The U.S. military has already spent $7.4 billion on AI to streamline and speed up video analysis in the conflict against ISIS.

The most promising AI effort the Pentagon has going now is Project Maven which started in July 2017 . Military analysts are using Google-developed AI algorithms to mine live video feeds from drones. The DoD is now developing an AI-driven algorithm to work in conjunction with its drone footage to spot, tag and bookmark potential threat targets. With machine learning techniques, software is taught to find particular objects or individuals at speeds that would be impossible for any human analyst. This AI technology can differentiate between people, objects and buildings, much like Google’s driverless cars. Undersecretary of Defense for Intelligence Joseph Kernan said Project Maven only started a year ago and so far has been “extraordinarily” useful in overseas operations.

Thousands of DoD intelligence analysts are currently employed to examine this video data, but it can take an entire team 24 hours to interpret only a fraction of one drone’s sensor data — a substantial roadblock to quickly catching terrorist suspects and preventing further attacks. The DoD is now developing an AI-driven algorithm to work in conjunction with its drone footage to spot, tag and bookmark potential threat targets. This AI technology can differentiate between people, objects and buildings, much like Google’s driverless cars.

In his latest Forbes article – titled Defense, Intelligence And The Role of Video In Counterterrorism – Linius CEO, Chris Richardson, explores the changing role of video in intelligence operations. Richardson argues that, while video is undoubtedly playing an increasingly important role in informing security and law enforcement agencies – at home and abroad, cumbersome analysis techniques and mushrooming video volumes are barriers to timely action.

Richardson goes on to say that manual footage interpretation and compilation processes, currently required to enable analysts and decision-makers to derive meaningful insights, meant that “reacting to events, rather than preventing them, is often the only possible course of action.”

AI and machine learning might have come a long way in assisting analysts and identifying pertinent aspects of video footage. But how do you then enable those different video segments to be quickly compiled into a single, watchable stream that authorities can understand and act on? And what about context? How can you compare selected footage against other video sources to glean additional insights and meaning? Until now, the answer has largely been time-consuming, human-intensive processes.

The Technical Limitations Of Digital Video

Currently, digital video files are singular, static blocks of data wrapped in a “container” — an MPEG or AVI format. Their inflexible nature severely limits the ability to quickly or easily analyze the specific contents of a video or compare one video source to another at a meaningful level. However, removing the container surrounding the video data and then applying the concept of virtualization to the exposed video data itself could enable two things:

The search, detection and comparison of specific objects tagged across any number of video sources (think faces, places, suspect actions, weapons or vehicles).

The instant assembly or reassembly of those identified objects into one seamless clip, without having to move videos from their separate secure locations.

Richardson says that applying video virtualization technology, with proven artificial intelligence capabilities, could empower “law enforcement and intelligence agencies to programmatically search across, and splice together, multiple video surveillance streams from disparate sources (such as drones, security cameras and archives).

“The upshot? The ability to systematically deliver the actionable intelligence needed to identify and mitigate threats in near real-time.”

Video Virtualization technologies

Melbourne-based Linius Technologies has released a publicly available platform that makes video searchable. Linius invented and patented a technology called Video Virtualization Engine (VVE), which makes the data inside an ordinary video searchable. In theory, if a Die Hard movie was converted using the technology, a viewer could search for all the times Bruce Willis yells “Yippee ki-yay, mother$#%er!”.

The technology has a variety of applications such as embedding personalised ads, searching and monitoring security footage and preventing piracy by locking a streaming movie to a viewer.

Google has a similar technology called Cloud Video Intelligence while Microsoft has a service called Video Indexer.

From today, Linius’s tech is available worldwide through a new program called Linius Video Services (LVS), a “Software as a Service” (or SaaS) platform.

Market Growth

Virtual video technology has the potential to revolutionize the way video is distributed and consumed around the globe. Video virtual technology has potential opportunities in four industries: Personalized Advertising, Anti-Piracy, Search, Security & Division

The global video virtualization market is projected to grow at a significant growth rate in the near future primarily due to growing video internet traffic which accounts for 80% of total internet traffic. Growth in the market will be primarily driven by growing demand of video surveillance, rising spending on personalized video advertisement, and rising demand for anti-piracy software.