Get all the news you need about Smart Buildings with the Memoori newsletter
Imagine trying to find things you like on the Internet without a search engine. It is a simple concept that highlights the importance of such tools. Without a search engine, you would literally have to read every page on the internet one-by-one until something interesting came up. It would take an unfathomable amount of time and you would inevitably miss a huge number of things you would have liked to find. In fact, web pages are probably created faster than you could even check them, essentially making the task impossible. That is similar to the state of video surveillance, without recent developments in artificial intelligence (AI).
Before AI video analytics we had people in front of screens. Usually one person in front of many screens attempting to simultaneously watch each screen looking for anything that might represent a security concern. An explosion or a disruption in a crowd of people may be fairly straightforward to spot but the vast majority of things that could raise an alarm would usually go unnoticed.
An unauthorized entry by non-violent means, a theft of an unsecured item, even a suspicious package placed in a trash can, and many other important situations would likely be missed by the human eye trying to scan multiple video streams. The cost of hiring enough security personnel to manage the growing number of cameras is simply too high. Even once an incident has taken place it remains incredibly difficult and time-consuming to comb through video recordings in order to collect evidence that would help security personnel identify suspects or understand the sequence of events.
“Current law enforcement systems are increasingly unable to cope with the sheer volume of surveillance material captured and stored every day. This is only set to rise, with the population of video cameras increasing by at least 12% per year. These video streams will only ever be useful if processes to search and analyze the mountain of data keep pace. As it stands today vital information is missed because the vast majority of video is simply never viewed,” explains our recent report: The Global Market for Intelligent Video Analytics 2018 to 2023.
AI video surveillance could change everything. A virtual security guard with an unlimited number of eyes and an endless attention span, giving it the capacity to track almost every action in every video stream 24 hours a day and seven days a week. When the AI spots something suspicious it can alert human security personnel so they can review the video, assess the risk and decide on the best course of action.
During forensic investigations, law enforcement and security personnel can utilize AI enabled search functions to track people and events in a fraction of the time and with far greater accuracy than would be possible with a human alone. All this isn’t as simple as just buying an AI system and giving it access to video streams, however, each AI needs to be trained and customized for their unique environment and set of circumstances.
The first part of implementing AI video surveillance, adding the analytics engines to the various video streams and enabling them, is relatively straightforward. The second part, configuring them for accurate performance, can be far more complex and time-consuming. Every site is unique, and even very similar sites using the same analytic rules may need to deliver very different results, based upon the specific operational requirements. Even with a self-learning AI system, there may be a need to adjust detection zones and masks, camera angles, perspective settings. It is this second part that has held back adoption.
“There has been much debate about exactly ‘how intelligent’ the technology really is and whether it provides satisfactory ROI. But in 2018, there is now a growing belief that video analytics could finally move beyond what has been achieved through conventional rule based systems,” our comprehensive report highlighted.
“This is due in large part to major advances in semiconductor architecture, which is enabling much faster processing; Empowering deep learning and machine learning algorithms to analyze data many times faster than was previously possible. Venture capitalists are now pouring billions of dollars into financing Artificial Intelligence (AI) chips and analytic software companies,” the report continued.
What is clear is that video surveillance systems currently generate such vast amounts of data that the information can simply not be utilized properly and that there is huge potential to maximize the value of this data through new technology. Recently developed chip architectures combined with AI video analytics software can be put to work on these huge volumes of data to vastly enhance the security, safety, and performance of people, buildings and the business enterprise. While we are still in the early stages of developing the technology, AI should have a central to the future of video surveillance.
“There is still much to be done in perfecting the technology and getting it to market, but these ‘new tools’ have opened up the opportunity to bring AI products to the video analytics market potentially revolutionizing its performance and capability,” our recent report explains. “If it can deliver, it will further drive demand for intelligent video surveillance, not just for new projects but open up a vast latent potential for retrofitting millions of existing camera installations.”