How Smart City Police Agencies Accelerate Investigations with Video Content Analytics

Going a step above legacy surveillance methods

Smart Cities use a variety of sensors to monitor activities and collect data in public spaces across a city, with the aim of being more effective and efficient in their management of traffic, waste, water resources, energy, and law enforcement. Those sensors usually include video surveillance resources that are useful for a variety of city agencies, including sanitation, transportation, planning and, of course, local law enforcement. Because video surveillance has been used for physical security for over 25 years, often the first step that a Smart City takes is to install — or expand upon — an extensive video camera network. Usually, the first agency to leverage the video camera footage is the city’s police department, so they can monitor key cameras in real time, and conduct forensic review of video footage during post-incident investigations.

For the sake of public safety, police departments need to resolve investigations quickly and accurately as possible, but it takes time to manually review the high volumes of video evidence, and human analysts are rarely able to effectively comprehend or analyze all the data that lies within it. Video analytics technology addresses both of those problems, by processing video data and identifying, classifying and indexing the objects in video footage – such as cars, trucks, buses, motorcycles, bicycles, women, men, children, and animals – to make video searchable, actionable and quantifiable.

Although VCA technology often facilitates real-time video monitoring with alerts to increase situational awareness and drive emergency response, its forensic capabilities can be very impactful for law enforcement. The technology enables police investigators to accelerate post-incident investigations by filtering video to rapidly review several hours — or even days — worth of video in a matter of minutes. By conducting searches and combining filters based on known case details, police can quickly and accurately sift through high volumes of video footage from across multiple cameras to gather key evidence and make better and faster decisions to quickly locate, identify and apprehend persons of interest.

The Jussie Smollet case in Chicago is only one example where police used VCA to search through video segments from hundreds of cameras – including a series of public and private cameras – to solve a crime. In that case the investigation was at first focused on finding the perpetrators of an alleged hate crime, but the video analysis ultimately led to charges against the complainant for filing a false police report and disorderly conduct.

Face recognition and appearance similarity

Police departments routinely use VCA to review street surveillance video footage when searching for a missing person, or person of interest. And, in jurisdictions where it is allowed, police departments can create and manage watchlists of digital images extracted from video or from external sources to conduct face matching within video. A video analytics system can also trigger real-time alerts whenever a face match is detected in a live camera view. Whether post-event or in real time, face recognition is a crucial tool for accelerating investigations and time-to-target. For jurisdictions that do not allow face recognition, police can search through video footage based on appearance similarity, filtering footage for individuals based on other features, such as clothing color and other attributes.

License plate recognition

A similar search and alert filter that is vital to law enforcement is “in the wild” license plate recognition, which detects license plates and registration numbers in video. Even a partial plate number can be useful to investigators, and they can search video feeds or manage watchlists and alerting based on available – event incomplete – information to find potential matches, investigate developing situations, and solve cases.

Police departments have no shortage of incidents to investigate, but they often lack enough manpower to investigate them all as fast as they would like to. Smart Cities can help police departments save valuable time, solve cases more accurately and quickly, and possibly prevent future crimes, by complementing video surveillance systems with video content analytics technology. It’s a sensible way to get more value from CCTV networks, increase police efficiencies, and improve public safety.