Artificial intelligence (AI) is improving everyday solutions, driving efficiency in ways we never imagined possible. From self-driving cars to intelligent analytics, the far-reaching impacts of Deep Learning-based technology empower human operators to achieve results more effectively while investing fewer resources and less time.
By introducing AI, solutions are not merely powered by data, but they also generate valuable intelligence. Systems which were once leveraged for a narrow, dedicated purpose, can suddenly be engaged broadly across an organisation, because the previously under-utilised data can be harnessed for enhancing productivity and performance.
When it comes to physical security, for instance, video surveillance is a standard solution.
Yet, by introducing AI-driven video analytics software, video data can be leveraged as intelligence in previously inaccessible ways. Here are some examples of how diverse organisations are using AI-based video intelligence solutions to enhance security and performance with searchable, actionable and quantifiable insights.
Law enforcement relies on video surveillance infrastructure for extracting investigation evidence and monitoring people and spaces. Instead of manual video review and live surveillance – which is prone to human error and distraction – police can harness video content analysis to accelerate video investigations, enhance situational awareness, streamline real-time response, identify suspicious individuals and recognise patterns and anomalies in video. The video intelligence software processes and analyses video to detect all the people and objects that appear; identify, extract and classify them; and then index them as metadata that can be searched and referenced.
For law enforcement, the ability to dynamically search video based on granular criteria is critical for filtering out irrelevant details and pinpointing objects of interest, such as suspicious persons or vehicles. Beyond accelerating video evidence review and extraction, police can leverage video analysis to configure sophisticated real-time alerts when people, vehicles or behaviours of interest are detected in video. Instead of actively monitoring video feeds, law enforcement can assess triggered alerts and decide how to respond. In this way, officers can also react faster to emergencies, threats and suspicious activity as it develops.
Empowering law enforcement to maintain public safety is important beyond the benefit of increasing security: A city with a reputation for effective, reliable law enforcement and enhanced safety is more likely to attract residents, visitors and new businesses, exponentially driving its economic development. Furthermore, in cities where law enforcement can work productively and quickly, time and human resources can be reallocated to fostering growth and building community.
Video analysis empowers cities to harness their video surveillance data as operational intelligence for optimising city management and infrastructure. When video data is aggregated over time, it can be visualised into dashboards, heatmaps and reports, so operators can identify patterns and more seamlessly detect anomalous behaviour.
A city could, for instance, analyse the most accident-prone local intersection and assess the traffic patterns to reveal details such as where cars are dwelling and pedestrians are walking; the directional flows of traffic; and the demographic segmentations of the objects detected: Are cars lingering in no-parking zones? Are pedestrians using designated crosswalks – is there a more logical location for the crosswalk or traffic light?
Do vehicles tend to make illegal turns – should police proactively deter this behaviour, or should the city plan new infrastructure that enables vehicles to safely perform these turns? Finally, does the rise in bike traffic warrant implementing dedicated biking lanes? With video intelligence, urban planners can answer these and other questions to facilitate local improvements and high quality of life.
Insight into traffic trends is also critical for transport companies, from public transit services to transportation hubs and airports. By leveraging the video insights about citywide traffic, public transit organisations can make data-driven decisions about scheduling and services. Analysing video surveillance around bus stops, for instance, can help these companies understand the specific hours per day people tend to dwell around bus stops. Correlating this information with transactional data for each bus line, bus schedules can be optimised based on demand for individual bus lines, shortening waiting times for the most popular routes.
Similarly, the traffic visualisations and activity heatmaps derived from the video of major transit hubs, such as international airports and central stations, can be beneficial for increasing security, enhancing situational awareness, identifying causes of congestion, improving throughput and efficiency and, ultimately, solving these inefficiencies to provide a streamlined customer experience for travellers.
Much like a city, large education campuses have internal transportation services, residential facilities, businesses and law enforcement, and video content analysis can support the campus in intelligently managing each of those business units, while also providing video intelligence to these individual groups.
Campus law enforcement can leverage video data to increase situational awareness and public safety, driving real-time responses with the ability to make informed assessments and accelerating post-event investigations with access to easily extractable video data.
When campuses are expanding or developing additional infrastructure, they can plan new crosswalks, traffic lights, roads, buildings and entrances and exits based on comprehensive video intelligence. By understanding where pedestrians and vehicles dwell, walk, cross or even violate traffic laws, the campus can inform construction projects and traffic optimisation.
Finally, the campus can leverage video business intelligence to justify leasing pricing for different retailers across campus, demonstrating property values based on traffic trends that can be correlated with retailer point of sale data. Whether its empowering security, productivity or decision-making, the insights generated by AI-based technology can drive significant optimisation – especially when data is fused and cross-referenced across smart sensors and systems for even deeper intelligence.
In the case of AI-backed video analytics, diverse organisations can harness video surveillance impactfully and dynamically. Whereas once video technology investments could be justified for their security value – with the introduction of AI capabilities – procurement teams can evaluate these solutions for countless business operations, because they offer broadly valuable intelligence. And video surveillance and analytics is merely one example of AI-driven solutions’ potential to disrupt business as we know it.