The business intelligence gathered through monitoring video surveillance streams in real-time using video analytics software, helps to identify patterns, attributes, and events of interest. Close human-based monitoring of many video feeds is simply not possible. With video analytics, the system can alert video operators of key events and automatically trigger specific actions or procedures.
Here are the most common types of video analytics:
Automatic License-Plate Recognition (ALPR) uses optical character recognition (OCR) technology on the captured video to identify and read vehicle license plates. Source videos can come from existing closed-circuit television (CCTV) or video cameras, law enforcement cameras, or high-speed ALPR cameras mounted on roadway infrastructures.
Face recognition (or more specifically, facial recognition) technology is capable of identifying and matching human faces from digital IP video. This data is typically compared against a database of authenticated users by matching unique facial features. Facial recognition video analytics can improve the building and perimeter security and the identification of known offenders. To prevent spoofing (such as holding a picture of a person up to the camera), spatial recognition (ensuring the image is not generated from a 2D source), as well as liveliness checks (the subtle movement of facial features), are typically deployed.
Crowd Detection is a real-time surveillance technology that allows for the detection of crowd density to evaluate capacity or occupancy issues within a defined area. Applications of crowd detection include population counting, public event management, disaster management, safety monitoring, military management, and suspicious activity detection.
The technology of people tracking requires the quantification of human behavioral motion and the detection of facial activities and attributes. Applications of people tracking include intrusion detection, wrong-way detection, people counting, physical distancing, and customer behavior analysis.
Left and Removed Item Detection is a video analytics technology based on monitoring the appearance and disappearance of static objects within a defined area. The technology is often deployed in public areas like airports or subways to detect potential bombs, although it can also be used to ensure fire escape areas remain clear and unobstructed.
Motion Tracking in video surveillance is designed to detect objects moving within a predefined area of interest reliably. Motion tracking impacts a wide range of industries, including the military, critical infrastructures, entertainment, sports, healthcare, and robotics.
Outdoor Object Tracking is a video analytic for detecting and monitoring the movement of vehicles and people in outdoor environments. Applications of object tracking include traffic control, visual surveillance, forensics, human-object interaction, gesture recognition, and augmented reality. New, AI-based object tracking analytics can identify the type of object, such as vehicle or person.
People Counters are video analytics optimized for indoor usage that counts the number of people passing through a specified area. Benefits can include occupancy control, capacity evaluation, sales & conversion metrics, personalized visitor experiences, and measuring operational effectiveness.
Pan-tilt-zoom (PTZ) auto-tracking video analytics enable PTZ CCTV surveillance cameras to follow and zoom in on people and vehicles within a field of view. Some benefits of PTZ Auto-Tracking include a larger field of view with fewer blind spots, motion tracking, and cost savings. Auto PTZ also enables an operator to monitor an event while leaving his or her hands-free to perform other tasks, such as using a telephone or engaging in other deterrent activities.
Video Motion Detection (VMD) is a feature that senses physical movement for a given area in real-time. Motion detection analytics are embedded in IP cameras, network video recorders, and video analytics, and video management software systems.