The ability to monitor and track moving objects, such as vehicles, people, assets, etc., has been possible ever since the introduction of PTZ cameras several decades ago. Using a joystick, mouse or a keyboard for manual control, system operators have been able to adjust cameras’ pan, tilt and zoom functionalities to better observe activity and to capture close-up detail for evidential purposes.
Moving on from manual real-time control, the use of pre-set tours has added to the value of PTZ-enabled cameras, as it means the devices can be programmed to patrol around a field of view. To some extent, this has minimised the amount of time an operator needed to efficiently monitor a site.
A significant improvement in how PTZ-enabled cameras efficiently monitored large open spaces occurred over a decade ago, when Hanwha Techwin and other technologically advanced manufacturers introduced PTZ dome cameras which offered an intelligent auto-tracking function. System operators could apply the function to specific objects, and the camera would then use its PTZ functionality to follow the object as it moved through the camera’s range of view.
However, the recent leap in processing capabilities and AI technology has allowed the effectiveness and efficiency of auto-tracking to advance to a new level. When an alarm event occurs, the latest generation of video surveillance cameras equipped with highly accurate Deep Learning AI-based video analytics can be triggered, locking onto people or vehicles which are potentially involved in criminal activity as they move around a camera’s field of view. No operator involvement is needed as the analytics enable the cameras to operate completely autonomously.
The accuracy of AI-based auto-tracking is enhanced as users are able to configure the technology to ignore video noise, swaying trees, moving clouds and animals. This makes it an extremely powerful hands-free tool to help operators monitor activity occurring in open environments such as airports, car parks, industrial estates, stadia and city centres.
Wisenet X PTZ PLUS Cameras
AI-based auto-tracking and an innovative hand-over feature have been incorporated into the latest Wisenet X PTZ PLUS cameras, making them an ideal video surveillance solution for large open areas and perimeter protection applications.
The PTZ PLUS cameras include hand-over triggers which switch adjacent and successive cameras to zoom into pre-set positions and then continue to auto-track the target person or vehicle as they move out of the field of view of the first camera. As with the AI-enabled tracking, this hand-over occurs autonomously without the need for operators to be involved, thus making control rooms more efficient. The function ensures operators who are not familiar with a site do not lose the location of a suspect, and minimises the risk that criminal activity will be unseen and unrecorded.
The 2MP, 6MP and 4K UHD cameras also feature precise manual PTZ control, improved pre-set accuracy, and an extended tilt range of up to 110 degrees which allows any objects positioned above the cameras to be seen. In addition, with the help of adaptive IR technology, built-in IR LEDs automatically focus on targets, enabling the cameras to capture high quality images of objects up to a distance of 200 metres away, regardless of the lighting conditions.
AI-based auto-tracking is designed to complement, and not replace, the expertise of control room operators. The ability for them to manually use PTZ cameras to proactively track what they believe may be suspicious activity and, if necessary, zoom in to view detail to verify what is occurring, remains in place. It is also not intended to use the AI-enabled capabilities to replace standard PTZ camera functionality, such as continuous pan and pre-set positioning. Both functions play an important role in deterring crime by creating a strong impression to offenders that they are being watched, even if the images captured by the cameras are not being monitored in real-time.
However, AI auto-tracking significantly adds to the ability of operators to multi-task without the danger of missing emerging or ongoing incidents. With research indicating it is difficult for anyone to focus for more than 20 minutes or so on scenes where there is little activity, the innovative technology can also help ensure operators do not fail to respond effectively to an incident because of a momentarily loss of concentration.